Methods and apparatus to determine a duration of media presentation based on tuning session duration
Methods, apparatus, systems, and articles of manufacture are disclosed to determine a duration of media presentation based on tuning session duration. Example apparatus a receiver to obtain a first tuning session duration indicative of an amount of time between channel changes of a first media presentation device at a first media presentation location, a presentation session estimator to select a model from storage, the model selected based on a match of the first tuning session duration and a second tuning session duration, the model including a relation between the second tuning session duration and a first presentation session duration of media presented on a second media presentation device at a second media presentation location, and estimate a second presentation session duration of media presented within the first tuning session duration based on the model.
1 . A method for generating enhanced-accuracy audience measurement data for a particular media, comprising:
receiving a first set of viewing data from a first plurality of set-top boxes displaying the particular media to a respective first plurality of televisions, the first set of viewing data comprising:
a first set of tuning session data indicative of a first set of tuning session durations; and
a first set of on-off session data indicative of a first set of on-off durations corresponding to the first set of tuning session durations;
generating, based on the first set of viewing data, a model for determining estimated presentation session durations from known tuning session durations, by:
dividing the first set of viewing data into a plurality of tuning session duration ranges;
for a tuning session duration range of the plurality of tuning session duration ranges:
determining a corresponding frequency distribution of on-off durations for each first viewing data entry with a tuning session duration in the tuning session duration range; and
calculating, from the corresponding frequency distribution of on-off durations, a corresponding expected presentation session duration for the tuning session duration range; and
combining the corresponding expected presentation session duration for at least a portion of the plurality of tuning session duration ranges to generate the model;
receiving a second set of viewing data from a second plurality of set-top boxes displaying the particular media on a respective second plurality of televisions, the second set of viewing data comprising:
a second set of tuning session data indicative of a second set of tuning session durations without corresponding on-off durations;
transforming the second set of viewing data into an enhanced second set of viewing data by:
for each second viewing data entry of the second set of viewing data:
using the model to determine an estimated presentation session duration from a turning session duration of each second viewing data entry; and
enhancing each second viewing data entry with the estimated presentation session duration; and
presenting, on a display, the enhanced-accuracy audience measurement data including the first set of viewing data and the enhanced second set of viewing data.
2 . The method of claim 1 ,
wherein generating the model further comprises:
for the tuning session duration range of the plurality of tuning session duration ranges:
determining a corresponding conditional distribution of the on-off durations for each first viewing data entry with the tuning session duration in the tuning session duration range; and
calculating, from the corresponding conditional distribution of the on-off durations, the corresponding expected presentation session duration for the tuning session duration range.
3 . The method of claim 1 ,
wherein the enhanced-accuracy audience measurement data is generated for a plurality of media using the model for the particular media.
4 . The method of claim 1 , further comprising:
receiving a third set of viewing data from a third plurality of set-top boxes displaying an additional particular media to a respective third plurality of televisions, the third set of viewing data comprising:
a third set of tuning session data indicative of a third set of tuning session durations; and
a third set of on-off session data indicative of a third set of on-off durations corresponding to the third set of tuning session durations;
generating, based on the third set of viewing data, an additional model for determining additional estimated presentation session durations from additional known tuning session durations, by:
dividing the third set of viewing data into a plurality of additional tuning session duration ranges;
for an additional tuning session duration range of the plurality of additional tuning session duration ranges:
determining a corresponding additional frequency distribution of additional on-off durations for each third viewing data entry with an additional tuning session duration in the additional tuning session duration range; and
calculating, from the corresponding additional frequency distribution of additional on-off durations, a corresponding additional expected presentation session duration for the additional tuning session duration range; and
combining the corresponding additional expected presentation session duration for at least a portion of the plurality of additional tuning session duration ranges to generate the additional model;
receiving a fourth set of viewing data from a fourth plurality of set-top boxes displaying the additional particular media on a respective fourth plurality of televisions, the fourth set of viewing data comprising:
a fourth set of tuning session data indicative of a fourth set of tuning session durations without corresponding additional on-off durations;
transforming the fourth set of viewing data into an enhanced fourth set of viewing data by:
for each fourth viewing data entry of the fourth set of viewing data:
using the additional model to determine an additional estimated presentation session duration from an additional turning session duration of each fourth viewing data entry; and
enhancing each fourth viewing data entry with the additional estimated presentation session duration; and
presenting, on the display, the enhanced-accuracy audience measurement data including the third set of viewing data and the enhanced fourth set of viewing data.
5 . The method of claim 1 ,
wherein generating the model further comprises:
for the tuning session duration range of the plurality of tuning session duration ranges:
determining a further corresponding frequency distribution of on-off durations at set viewing times for each first viewing data entry with the tuning session duration in the tuning session duration range.
6 . The method of claim 5 , further comprising:
generating, based on the first set of viewing data, an additional model for determining the estimated presentation session durations from the known tuning session durations, by:
for the tuning session duration range of the plurality of tuning session duration ranges:
determining a corresponding conditional distribution of the on-off durations at the set viewing times for each first viewing data entry with the tuning session duration in the tuning session duration range; and
calculating, from the corresponding conditional distribution of the on-off durations at the set viewing times, the corresponding expected presentation session duration for the tuning session duration range; and
combining the corresponding expected presentation session duration for at least a portion of the plurality of tuning session duration ranges to generate the additional model.
7 . The method of claim 1 ,
wherein calculating the corresponding expected presentation session duration for the tuning session duration range comprises:
determining a weighted average across the corresponding frequency distribution of the on-off durations.
8 . The method of claim 1 , further comprising:
for each second viewing data entry of the second set of viewing data:
using the model to determine a standard deviation associated with a second tuning session duration of each second viewing data entry; and
enhancing each second viewing data entry with the standard deviation.
9 . The method of claim 1 , further comprising:
for at least one second viewing data entry of the second set of viewing data:
determining a weighted average of a set of grouped frequency distributions to determine the estimated presentation session duration.
10 . The method of claim 1 , further comprising:
for each second viewing data entry of the second set of viewing data:
determining a number of total viewing data entries for the corresponding frequency distribution from the model;
comparing the number of total viewing data entries to a predetermined entry threshold; and
if the number of total viewing data entries is less than the predetermined entry threshold, grouping one or more frequency distributions to determine the estimated presentation session duration.
11 . A non-transitory computer-readable storage medium, having stored thereon program instructions that, upon execution by a processor, cause performance of a set of operations for generating enhanced-accuracy audience measurement data for a particular media, the set of operations comprising:
receiving a first set of viewing data from a first plurality of set-top boxes displaying the particular media to a respective first plurality of televisions, the first set of viewing data comprising:
a first set of tuning session data indicative of a first set of tuning session durations; and
a first set of on-off session data indicative of a first set of on-off durations corresponding to the first set of tuning session durations;
generating, based on the first set of viewing data, a model for determining estimated presentation session durations from known tuning session durations;
receiving a second set of viewing data from a second plurality of set-top boxes displaying the particular media on a respective second plurality of televisions, the second set of viewing data comprising:
a second set of tuning session data indicative of a second set of tuning session durations without corresponding on-off durations;
transforming the second set of viewing data into an enhanced second set of viewing data by:
for each second viewing data entry of the second set of viewing data:
using the model to determine an estimated presentation session duration from a turning session duration of each second viewing data entry; and
enhancing each second viewing data entry with the estimated presentation session duration; and
presenting, on a display, the enhanced-accuracy audience measurement data including the first set of viewing data and the enhanced second set of viewing data.
12 . The non-transitory computer-readable storage medium of claim 11 , wherein generating the model is performed by:
dividing the first set of viewing data into a plurality of tuning session duration ranges;
for a tuning session duration range of the plurality of tuning session duration ranges, calculating a corresponding expected presentation session duration for the tuning session duration range; and
combining the corresponding expected presentation session duration for at least a portion of the plurality of tuning session duration ranges to generate the model.
13 . The non-transitory computer-readable storage medium of claim 12 , wherein calculating the corresponding expected presentation session duration for the tuning session duration range comprises:
determining a corresponding frequency distribution of on-off durations for each first viewing data entry with a tuning session duration in the tuning session duration range; and
calculating, from the corresponding frequency distribution of on-off durations, the corresponding expected presentation session duration for the tuning session duration range.
14 . The non-transitory computer-readable storage medium of claim 13 , wherein calculating the corresponding expected presentation session duration for the tuning session duration range further comprises:
determining a weighted average across the corresponding frequency distribution of the on-off durations.
15 . The non-transitory computer-readable storage medium of claim 12 , wherein calculating the corresponding expected presentation session duration for the tuning session duration range comprises:
for the tuning session duration range of the plurality of tuning session duration ranges:
determining a corresponding conditional distribution of the on-off durations for each first viewing data entry with the tuning session duration in the tuning session duration range; and
calculating, from the corresponding conditional distribution of the on-off durations, the corresponding expected presentation session duration for the tuning session duration range.
16 . The non-transitory computer-readable storage medium of claim 12 , wherein calculating the corresponding expected presentation session duration for the tuning session duration range comprises:
for the tuning session duration range of the plurality of tuning session duration ranges:
determining a corresponding frequency distribution of on-off durations at set viewing times for each first viewing data entry with the tuning session duration in the tuning session duration range; and
calculating, from the corresponding frequency distribution of on-off durations, the corresponding expected presentation session duration for the tuning session duration range.
17 . The non-transitory computer-readable storage medium of claim 11 ,
wherein the enhanced-accuracy audience measurement data is generated for a plurality of media using the model for the particular media.
18 . A computing system comprising:
a processor; and
a non-transitory computer-readable storage medium, having stored thereon program instructions that, upon execution by the processor, cause performance of a set of operations comprising:
receiving a first set of viewing data from a first plurality of set-top boxes displaying a particular media to a respective first plurality of televisions, the first set of viewing data comprising:
a first set of tuning session data indicative of a first set of tuning session durations; and
a first set of on-off session data indicative of a first set of on-off durations corresponding to the first set of tuning session durations;
generating, based on the first set of viewing data, a model for determining estimated presentation session durations from known tuning session durations;
receiving a second set of viewing data from a second plurality of set-top boxes displaying the particular media on a respective second plurality of televisions, the second set of viewing data comprising:
a second set of tuning session data indicative of a second set of tuning session durations without corresponding on-off durations;
transforming the second set of viewing data into an enhanced second set of viewing data by:
for each second viewing data entry of the second set of viewing data:
using the model to determine an expected presentation session duration from a turning session duration of each second viewing data entry; and
enhancing each second viewing data entry with the expected presentation session duration; and
presenting, on a display, enhanced-accuracy audience measurement data including the first set of viewing data and the enhanced second set of viewing data.
19 . The computing system of claim 18 , wherein calculating the expected presentation session duration comprises:
determining a corresponding frequency distribution of on-off durations for each first viewing data entry with a tuning session duration in each of a plurality of tuning session duration ranges; and
calculating, from the corresponding frequency distribution of on-off durations, a corresponding expected presentation session duration.
20 . The computing system of claim 18 , wherein calculating the expected presentation session duration comprises:
for a tuning session duration range of a plurality of tuning session duration ranges:
determining a corresponding conditional distribution of on-off durations for each first viewing data entry with the tuning session duration in the tuning session duration range; and
calculating, from the corresponding conditional distribution of the on-off durations, a corresponding expected presentation session duration for the tuning session duration range.