Methods, systems, articles of manufacture, and apparatus to estimate audience population
Methods, apparatus, systems and articles of manufacture are disclosed to estimate an audience population. An example apparatus includes at least one memory; instructions in the apparatus; and processor circuitry to execute the instructions to: determine whether respective ones of respondents are associated with a characteristic; detect unique instances of the respective ones of the respondents; in response to the respective ones of the respondents being associated with the characteristic, increase a sample capture count by one; in response to detecting the unique instances of the respective ones of the respondents exhibiting the characteristic, increase a unique capture count by one; determine a seed population estimate based on the unique capture count; and determine a population estimate having the characteristic based on the sample capture count, the unique capture count, the seed population estimate, and a number of available samples.
1 . An audience population subset size estimation computing system comprising:
memory having stored therein computer readable instructions;
at least one processor configured to, upon execution of the computer readable instructions, cause performance of a set of operations comprising:
identifying ones of a plurality of respondents included in an audience sample that are associated with a particular characteristic corresponding to an audience population subset;
determining a first sample quantity of the identified ones of the plurality of respondents;
determining a second unique sample quantity of the identified ones of the plurality of respondents having respondent identifiers unique amongst a set of stored respondent identifiers;
incrementing a third unique capture quantity by the second unique sample quantity;
determining a seed population estimate based on the third unique capture quantity, wherein the seed population estimate is greater than the third unique capture quantity; and
estimating the audience population subset size based on the first sample quantity, the third unique capture quantity, and the seed population estimate.
2 . The audience population subset size estimation computing system of claim 1 , wherein the operations further include:
accessing respondent identifiers of the identified ones of the plurality of respondents; and
updating the set of stored respondent identifiers to include the identified ones of the plurality of respondents having respondent identifiers unique amongst the set of stored respondent identifiers.
3 . The audience population subset size estimation computing system of claim 2 , wherein the respondent identifiers include at least one of an internet cookie, a MAC address, or an IP address.
4 . The audience population subset size estimation computing system of claim 1 , wherein the operations further include estimating a recapture probability of the respondents.
5 . The audience population subset size estimation computing system of claim 4 , wherein the recapture probability of the respondents varies with time.
6 . A method for estimating an audience population subset size, the method comprising:
identifying ones of a plurality of respondents included in an audience sample that are associated with a particular characteristic corresponding to an audience population subset;
determining a first sample quantity of the identified ones of the plurality of respondents;
determining a second unique sample quantity of the identified ones of the plurality of respondents having respondent identifiers unique amongst a set of stored respondent identifiers;
incrementing a third unique capture quantity by the second unique sample quantity;
determining a seed population estimate based on the third unique capture quantity, wherein the seed population estimate is greater than the third unique capture quantity; and
estimating the audience population subset size based on the first sample quantity, the third unique capture quantity, and the seed population estimate.
7 . The method of claim 6 , further comprising:
accessing respondent identifiers of the identified ones of the plurality of respondents; and
updating the set of stored respondent identifiers to include the identified ones of the plurality of respondents having respondent identifiers unique amongst the set of stored respondent identifiers.
8 . The method of claim 7 , wherein the respondent identifiers include at least one of an internet cookie, a MAC address, or an IP address.
9 . The method of claim 6 , further comprising estimating a recapture probability of the respondents.
10 . The method of claim 9 , wherein the recapture probability of the respondents varies with time.
11 . A non-transitory computer readable storage medium comprising instructions that, when executed, cause at least one processor to cause performance of:
identifying ones of a plurality of respondents included in an audience sample that are associated with a particular characteristic corresponding to an audience population subset;
determining a first sample quantity of the identified ones of the plurality of respondents;
determining a second unique sample quantity of the identified ones of the plurality of respondents having respondent identifiers unique amongst a set of stored respondent identifiers;
incrementing a third unique capture quantity by the second unique sample quantity;
determining a seed population estimate based on the third unique capture quantity, wherein the seed population estimate is greater than the third unique capture quantity; and
estimating an audience population subset size based on the first sample quantity, the third unique capture quantity, and the seed population estimate.
12 . The non-transitory computer readable storage medium of claim 11 , wherein the instructions, when executed, further cause the at least one processor to cause performance of:
accessing respondent identifiers of the identified ones of the plurality of respondents; and
updating the set of stored respondent identifiers to include the identified ones of the plurality of respondents having respondent identifiers unique amongst the set of stored respondent identifiers.
13 . The non-transitory computer readable storage medium of claim 11 , wherein the instructions, when executed, further cause the at least one processor to cause performance of estimating a recapture probability of the respondents.
14 . The non-transitory computer readable storage medium of claim 13 , wherein the recapture probability of the respondents varies with time.