Methods and apparatus to correct age misattribution in media impressions

Abstract

This disclosure relates generally to audience measurement, and, more particularly, to correcting age misattribution in media impressions. An example method involves predicting a first age probability density function for an audience member at a first time, the first age probability density function indicative of a first probability that the audience member is in a first age range at the first time. The example method also involves, in response to receiving media monitoring data associated with the subscriber at a second time after the first time, determining a second age probability density function by applying a first aging factor to the first age probability density function for the audience member, the second age probability density function indicative of a second probability that the audience member is in the first age range at the second time, and associating the media monitoring data with the second age probability density function.

What is claimed is: 
     
         1 . A method comprising:
 predicting a first age probability density function for an audience member at a first time, the first age probability density function indicative of a first probability that the audience member is in a first age range at the first time; and   in response to receiving media monitoring data associated with the audience member at a second time after the first time:
 determining a second age probability density function by applying a first aging factor to the first age probability density function for the audience member, the second age probability density function indicative of a second probability that the audience member is in the first age range at the second time, and 
 associating the media monitoring data with the second age probability density function. 
   
     
     
         2 . A method as defined in  claim 1 , further comprising determining the first aging factor by dividing a first number of days between the first time and the second time by a second number of total days in the first age range. 
     
     
         3 . A method as defined in  claim 1 , wherein the first age probability density function is indicative of a third probability that the audience member is in a second age range at the first time, and further comprising in response to receiving the media monitoring data associated with the audience member at the second time after the first time:
 determining the second age probability density function by applying a second aging factor to the third probability indicated by the first age probability density function for the audience member to create a fourth probability that the audience member is in the second age range at the second time.   
     
     
         4 . The method of  claim 3 , wherein the fourth probability comprises a portion of the first probability and a portion of the third probability. 
     
     
         5 . The method of  claim 1 , wherein the audience member is a subscriber to a database proprietor, and further comprising predicting the first age probability density function for the audience member further using an age prediction model based on subscriber characteristics of the audience member collected by the database proprietor. 
     
     
         6 . The method of  claim 5 , wherein the age prediction model is a classification tree and the first age probability density function is a terminal node of the classification tree. 
     
     
         7 . An apparatus, comprising:
 an age predictor to predict a first age probability density function for an audience member at a first time, the first age probability density function indicative of a first probability that the audience member is in a first age range at the first time; and   an age updater to, in response to receiving media monitoring data associated with the audience member at a second time after the first time:
 determine a second age probability density function by applying a first aging factor to the predicted first age probability density function for the audience member, the second age probability density function indicative of a second probability that the audience member is in the first age range at the second time, and 
 associate the media monitoring data with the second age probability density function. 
   
     
     
         8 . An apparatus as defined in  claim 7 , wherein in the age updater is further to determine the first aging factor by dividing a first number of days between the first time and the second time by a second number of total days in the first age range. 
     
     
         9 . An apparatus as defined in  claim 7 , wherein the first age probability density function is indicative of a third likelihood that the audience member is in a second age range at the first time, and wherein in response to receiving media monitoring data associated with the subscriber at the second time after the first time, the age updater is further to:
 determine the second age probability density function by applying a second aging factor to the third probability of the first age probability density function for the audience member to create a fourth probability that the audience member is in the second age range at the second time.   
     
     
         10 . An apparatus as defined in  claim 9 , wherein the fourth probability comprises a portion of the first probability and a portion of the third probability. 
     
     
         11 . The apparatus of  claim 7 , wherein the audience member is a subscriber to a database proprietor, and the age predictor is to predict the first age probability density function for the audience member fusing an age prediction model based on subscriber characteristics of the audience member collected by the database proprietor. 
     
     
         12 . The apparatus of  claim 11 , wherein the age prediction model is a classification tree and the first age probability density function is a terminal node of the classification tree. 
     
     
         13 . A tangible computer readable storage medium comprising instructions which, when executed, cause a machine to at least:
 predict a first age probability density function for an audience member at a first time, the first age probability density function indicative of a first probability that the audience member is in a first age range at the first time; and   in response to receiving media monitoring data associated with the subscriber at a second time after the first time:
 determine a second age probability density function by applying a first aging factor to the first age probability density function for the audience member, the second age probability density function indicative of a second probability that the audience member is in the first age range at the second time, and 
 associate the media monitoring data with the second age probability density function. 
   
     
     
         14 . A method as defined in  claim 13 , further comprising determining the first aging factor by dividing a first number of days between the first time and the second time by a second number of total days in the first age range. 
     
     
         15 . A method as defined in  claim 13 , wherein the first age probability density function is indicative of a third probability that the audience member is in a second age range at the first time, and further comprising in response to receiving the media monitoring data associated with the audience member at the second time after the first time:
 determining the second age probability density function by applying a second aging factor to the third probability indicated by first age probability density function for the subscriber to create a fourth probability that the subscriber is in the second age range at the second time.   
     
     
         16 . The tangible computer readable storage medium of  claim 15 , wherein the fourth probability comprises a portion of the first probability and a portion of the third probability. 
     
     
         17 . The tangible computer readable storage medium of  claim 13 , wherein the audience member is a subscriber to a database proprietor, wherein the instructions, when executed are further to cause the machine to predict the first age probability density function for the audience member using an age prediction model based on subscriber characteristics of the audience member collected by the database proprietor. 
     
     
         18 . The tangible computer readable storage medium of  claim 17 , wherein the age prediction model is a classification tree and the first age probability density function is a terminal node of the classification tree.