The problems with both parties’ convention viewership claims

After Nielsen reported higher traditional TV viewership of the 2020 Democratic National Convention than the Republicans‘, Donald Trump took to Twitter claiming the RNC actually beat the Democrats if you include and digital — though how do those claims stack up?

Nielsen’s data estimates that 23.4 million tuned in to watch Trump’s speech on the final night of the convention via linear TV, lower than ’s 24.6 million.

That Nielsen data does not include (nor did it ever claim to) account for the various digital ways that people could have watched the event.

Trump, meanwhile, claimed that he had 147.9 million total views during the entire week, while the Democrats were widely reported as having 122 million viewers during their turn.

That said, it’s important to look at the labels attached to each of those numbers. Notice how Trump’s reads “views” and the DNC’s read “viewers”? Those two things can be quite different.

Most ratings systems attempt to, in at least one of its metrics, estimate the number of people who watched something (as opposed to only counting everyone in a house of, say, five, as one).

Under this terminology, the same person is generally not considered a new “viewer” if they turn off the TV and then return later, though this does vary depending on methodology used. Some systems may put a time limit of, for example, 30 minutes or an hour where the viewer gets counted again if they leave and return.

“Views,” meanwhile, is a much trickier metric to define — because there’s no standard way to define what a “view” actually is.

The data could indicate everything from the number of times the stream was started (including by the same person potentially dozens of times) to how many people watched, say, at least 30 seconds of the stream (and again, people could be counted more than once if they left and then returned later).

The data reported for both the DNC and RNC combined traditional TV and digital and therefore includes “self reported” data from the parties and campaigns for the latter.

Because of that, it’s not clear just how accurate the data from either party is — the parties would likely need to hire an outside auditor with access to raw data to validate their findings in order for the numbers to be considered more accurate, at least in the eyes of an independent observer.

In most cases, the Democrats labeled their data as “viewers,” while Trump and the Republicans cited “views” in news releases and statements that were then digested into thousands of news stories — though some reporting appears to use the terms interchangeably, which could be widely inaccurate.

“Views” can have a much broader definition and potentially means that the same person was counted multiple times each time they returned and watched a significant portion of the event. In that sense it might not represent the true “reach” of the event.

It’s important to note that it’s not immediately clear exactly how either campaign defines “views” or “viewers” (and again, this is relying on self reported data to begin with) so it’s tough to get a clear picture of how many people actually watched either party’s proceedings online.

In theory, both campaigns could be using the same criteria but just calling the metric two different names, though the fairly consistent use of different terminology could be significant.