Weighted vs. Unweighted
Nielsen Media Research
included sample weights in their PxP data
starting in September 2003.
Prior to that date, the sample was
"self-weighted", meaning that the sample represented the
population for all demographic characteristics. In September, Nielsen
began merging Local People Meter (LPM) sample households into the national
sample. There are proportionally more LPM households in the LPM markets than
there are national households in the national sample, thus requiring the
introduction of weights to balance the
What is the difference between the
Weighted and Unweighted
- Weighted - The weights
measure the number of people in the population that are represented by each
member of the sample. For example, if sample member has a weight of 20,000 for
a selected day, this means that on that day the sample member represents
20,000 in the population.
- Unweighted -
If Unweighted is selected, then all analyses will
function based on the unweighted sample. In
other words, sample members will be weighted equally. Each sample member will
effectively be assigned a weight of 1.
When should I use the Weighted sample?
As a rule of thumb, use the Weighted sample.
Weighted sample analysis, which is the default, is more accurate in representing the
population, and should generally be used.
When should I use the
Unweighted sample analysis could be used if:
- You want to recreate an analysis that was made in Audience Watch version
5.73 or earlier.
- You really do want to see the viewing behavior of
individual sample members.
How do I know which sample I am using?
A Weighted (Wgt) and Unweighted (UnWgt) indicator
appears in the demographic description line to indicate whether analyses
uses the Nielsen Weighted or Unweighted sample.
FOR EXAMPLE, below is the demographic description line for
a Weighted (left) and Unweighted (right) sample:
Toggle between the Weighted and Unweighted sample
- Left-Click on the
Demographic Settings button on the
- From the
Age/Gender tab select Weighted or Unweighted (see image to