One quarter of the GorDaq index is defined by the quality of the Chancellor's media coverage. But how can such a thing be defined scientifically?
Newsnight has commissioned media analysts Romeike to do the job.
So how do they do it?
Public relations of this kind is at best semi-controlled since the journalist ultimately determines what will appear.
As the communications move from the "controlled" arena to the "uncontrolled," credibility increases.
It is the role of the journalist which dictates PR's "semi-controllability" and, as such, its credibility and then, ultimately, its GorDaq rating.
Any attempt to create a rating would have to reflect the Quality of the write-up along with Frequency (how often it gets repeated) and Reach (how many see it).
We knew from the beginning that the formula would have to look something like this: (Frequency) x (Reach) x (Quality).
Frequency and reach are fairly objective standards and readily defined using long standing market research measures.
The challenge became "how does one define 'quality'" and then "how does one distil these three measures into a single number?"
Our analysts began by creating a list of possible attributes thought to indicate "quality" in a news story.
The list contained over 40 attributes, such as "a story reporting news exclusively about one individual" rather than "a round-up story mentioning several individuals," for example.
A representative sample of coverage was gathered from a variety of news sources. Stories were selected and then coded by expert media analysts.
Some attributes were rated on a scale of one to five, (where, for example, a (1) would indicate "a round-up story mentioning several individuals" and a (5) would indicate "a story reporting news exclusively about one individual").
Other attributes were binary, in that the attribute was either present or not (the presence of a visual photo or graphic, for example) representing the individual.
Each of the attributes was scored for each of the articles. The tone of the article was also assessed, so that a "an exclusive story with a photo" in a positive article was good just as "an exclusive story with a photo" in a negative article was bad.
The second phase was designed to reveal the extent to which each of the possible attributes drove awareness and recall among the audience.
Copies of each article were distributed in a controlled environment. Participants read them and were later asked the extent to which they remembered specific stories and story themes.
Certain stories emerged as the most memorable, and these stories each contained common story attributes which were key drivers of awareness and recall. These factors form the basis for Impact Score & Net Effect measure of media performance.
While the exact formula is a closely guarded secret, the factors include:
News item placementHeadlineInitial mentionExtent of mentionDominanceVisuals
Impact Score is a composite of Prominence and Tone that becomes a "quality" score on a scale of +100% to -100%. (Tone determines whether the "score" is a positive or negative number).
Prominence is a research-based estimation of the likelihood that someone will be exposed to and remember news about an individual.
Prominence factors are weighted based on their relative contribution to news item "quality."
Net Effect is the result of the Reach of an individual news item multiplied by its Impact Score. Results are combined to determine an overall Net Effect for a specific period of time, or surrounding a particular issue or event. Net Effect can be positive or negative.