[ghaff]

One thing mentioned--but largely passed over--in the article is the following:

Average writer earnings in 2021: $260,000.

Average earnings do not account for the sizable number of WGA members who, in any given year, earn nothing from writing. The WGA has not released a median annual figure since 2014, when it was $140,000 (in 2021 dollars, for the sake of consistency).

I'm not actually surprised. So writing is actually a pretty good gig (even if you're not a showrunner/executive producer) if you have a real gig. But the median WGA member is probably waiting tables or has some other job. And that's WGA members. The typical writer submitting scripts to studios on spec is earning almost nothing.

ADDED: And this is in the context of writing, perhaps other than some corporate content job, being an increasingly terrible way to earn a living on average.

[lotsofpulp]

Anytime I read a single statistic without the accompanying distribution in deciles, or at least quintiles, I assume there is an agenda or the publisher is trying to hide something.

[gruez]

Seriously? Isn't that an overly cynical worldview? I agree that having distribution information is strictly better than only the median, but in most cases only the median/mean is mentioned for the sake of brevity. People reading a article about writer pay don't want 10 numbers thrown at them for a single statistic.

[lotsofpulp]

It is just my experience over 20 years of reading. A 2x10 or 2x5 table is not going to make an article too long.

But all too often, the word “average” will be used without even clarifying if it is mean or median, and I have seen too many cases where it just so happens to support whatever argument the publisher is making.

See even the example ghaff quoted. The union went out of its way to remove information about the earnings distribution because they did not want people to know how the median was moving.

[gruez]

> A table with 2x10 table with the data is not going to make an article too long

It is if you're going to do it for every statistic. eg. in a story about gen Z skipping college, you could have statistics about tuition rate, student loan amount, time to graduate, graduation rate, earnings after graduation, etc. Add to that, all of the statistics but separated by various demographic factors (eg. geography, age, race), and you can easily have dozens of tables. That's fine if you're writing a 20 page graduate dissertation, but for a daily article on npr.org or whatever that's just overkill.

Not to mention, in many cases the reason why it's being omitted from news stories isn't because of some nefarious motives by the author, it's because the source material only mentions medians. eg. most of the BLS news releases only has medians/averages: https://www.bls.gov/news.release/empsit.toc.htm, so any news stories based off them are inevitably going to not have decile level data like you demand, through no nefarious motivation on the part of the writer.

[lotsofpulp]

> That's fine if you're writing a 20 page graduate dissertation, but for a daily article on npr.org or whatever that's just overkill.

If the publisher is taking on a complicated topic, then it does need all of that information, at least via a link. Otherwise, we have the never ending clickbait of bullshit “articles” with one nebulous average intending to lead people to think something notable has happened, when it really has not.

There is so much propaganda floating around because people accept massaged averages as truth, I cannot imagine it is a benefit to society. Also, in this context, we happen to have wonderful machines and networking that allow us to share data at basically zero marginal cost.

> Not to mention, in many cases the reason why it's being omitted from news stories isn't because of some nefarious motives by the author, it's because the source material only mentions medians.

Yes, it goes without saying that the blame them moves up to the entity that has the data, but chooses not to release it.

[ghaff]

In this case, the distribution (for which it's not clear the numbers are public) is pretty important. A storyline that showrunners of hit shows make bank while the typical scriptwriter sending a spec script in to a studio makes zilch is basically "water is wet." Judging whether writers in film/TV in general have a real gripe with compensation (given that median salaries seem pretty decent) needs some more granularity.

[ghaff]

In all fairness, I'm not sure a population that includes everyone who has ever written a script and sent it to a studio is a useful study point for most purposes. (I think we know the answer. Very few make it in any reasonable way.) But a distribution of those who have made some income above a reasonable threshold over the past 5 years is probably useful if we want to understand the situation for people legitimately working in the profession.

[lotsofpulp]

Right, but it is not in the writer’s union to let everyone understand the true situation, presumably because it would hamper their negotiating position.

Not that the union has any obligation to release the data, but just an example for why I assume I am not getting a clear picture when I am not provided the distribution, when it would be trivial to do so.