Facebook’s ad transparency tool – the Facebook Ad Library – is a good first effort, and any progress to make political advertising more transparent should be encouraged and commended, but as we and others have noted, there is much still that can be done to make it more useful.
In an effort to make the political ads data provided by Facebook more useful and accessible, we recently updated PoliDashboard, a publicly accessible, data visualization tool developed by our lab to help voters, journalists and campaign staffers monitor political discourse in Canada. One of the main features of PoliDashboard is the Facebook Political Ads Module which shows information about both active and inactive ads about social issues, elections or politics across Facebook products in Canada. The data is automatically updated every four hours via the Facebook Ad API.
The key new insight that Polidasboard offers over Facebook’s own Ad Library is that it automatically aggregates political ads purchased by an advertiser, giving users a bird’s eye view of how individual advertisers in Canada deploy their ad budget. Users can click on any advertiser shown on the ads module which would take them to interactive charts listing the ads the advertiser ran, who was shown the ads to and where the ads were shown during a given period of time.
What the Facebook Ad Library Displays to Users
For every ad on social issues or politics, as identified by Facebook or self-declared by the advertiser, the Facebook Ad Library provides stats on the percentages of Facebook users who were shown the ad and where it was shown (broken out by the users’ age, gender and region).
For example, Figure 1 shows a sample ad targeting seniors from the Conservative Party of Canada (CPC) as displayed by the Facebook Ads Library. If you look at the “Who Was Shown This Ad” chart, you can quickly determine that at 21% of the total audience for this ad, men 65+ were the largest group of Canadians shown this ad to.
Seeing the Forrest for the Trees
If you are only interested in knowing about the reach of just this one ad, this chart as provided by Facebook is useful. But it is not very helpful if you are trying to understand the overall ads targeting and communication strategy of an advertiser over time.
To give voters a bird’s eye view of how individual advertisers, such as political parties, spend their ad budget on Facebook as a whole and who they are targeting with their various ads, we automatically aggregate gender and age breakdowns for all ads purchased by an individual advertise into a single chart and summarize them using box plots as shown in Figure 2. We do the same for the breakdown based on provinces (see Figure 3).
Each dot in these charts represents a single ad. Hover over any dot on the chart to see which other demographic group(s) on Facebook were shown that particular political ad (labelled as A in Figure 4). If all other dots that appear on the chart (other than the dot you hovered over) is at or close to the 0% mark on the x-axis (labelled as B in Figure 4), it suggests that particular ad was only (or mostly) targeted at that specific demographic group and no other. Users can also click on the dot to see the actual ad and all of its stats as provided by Facebook Ad Library in a new window.
How to Interpret a Box Plot Chart
If you are new to box plots, check out this explainer post. In short, a box plot is a convenient way to summarize a series of data points (a dataset). The beginning and end of the line indicate the min and max value respectively, excluding outliers (Figure 5). Data points displayed before or after the min and max values are considered to be statistical outliers. In our case, outliers are one-off ads that tend to target a particular demographic group more or less than usual.
The “box” in a box plot represents where 50% of all data points reside. The vertical line inside the box represents the median value, often defined as the “middle” value because it divides the lower half from the higher half of values in the dataset. In our context, the median value can be considered as a “typical” percentage of that demographic group targeted by the advertiser based on a given time period.
