EppsNet Archive: Bias

Big Losers

27 May 2017 /

I saw this headline on an AP story today — Poor and disabled big losers in Trump budget.

The story includes a photo of the budget (see below), so I think it’s safe to say that the AP writer didn’t read the entire thing before announcing who the “big losers” are. He’s just flogging his own agenda. (See also Harvard Study Says Media Are Very Biased Against Donald Trump)

2018 budget

“Trump’s plan for the budget year beginning Oct. 1 makes deep cuts in safety net programs . . .” the story says.

What’s the difference between a “cut” and a “deep cut”? The latter sounds mean and scary. Why not just say something factual like “10 percent cut” or “50 percent cut” and let readers put their own characterization on it?

“Safety net programs” is also a loaded expression.

“Trump’s budget would cut the food stamp program by $191 billion over the next decade.” OK, there’s a factual assertion. But the government doesn’t use zero-based budgeting, in which each year’s budget starts at zero and all expenses must be justified for each new period. Government budgeting calls for incremental increases over previous budgets.

For example, if the cost of Program X is budgeted to increase 10 percent per year, and we “cut” it by 5 percent, the cost still goes up 5 percent. We can have budget cuts and more spending at the same time.

So “Trump’s budget would cut the food stamp program by $191 billion over the next decade” actually means something like — I don’t know the actual numbers, but something like “The cost of the food stamp program would have gone up by $800 billion over the next decade but because of the $191 billion ‘cut,’ it will only go up by $609 billion.”

Also — numbers in a federal budget look really big because the US is a big country with a lot of people — 320 million. If you wanted to give every person in the country 5 dollars, you’d need to have more than $1.5 billion on hand to do it.

A small number — 5 bucks — becomes a big number — $1.5 billion — when you project it to a national scale.

Food stamp costs of $191 billion over a decade comes to $19 billion per year. How many people receive food stamps? About 45 million. So $191 billion is only about $400 per person per year.

I saw Elizabeth Warren on YouTube emoting about the budget: Blah blah blah Donald Trump blah blah blah Betsy DeVos blah blah blah …

I could be wrong but I don’t think Elizabeth Warren or fans of Elizabeth Warren really care about people on food stamps, at least not enough to help out of their own pockets. I think they care about having the power to deem things worthy and then make other people pay for them.

Elizabeth Warren didn’t say “I am personally contributing $400 per year and I want all of you who are as outraged about this as I am to contribute $400 per year to make up the difference in the food stamp budget.” What would happen if she did?


Harvard Study Says Media Are Very Biased Against Donald Trump

23 May 2017 /

According to a Harvard University study, the mainstream media are very biased against Donald Trump.

Here’s a chart from the study, showing that the tone of some news outlets is negative in as many as 98% of reports:

Tone of Trump news coverage

I’ve noticed that even our local news station is about 90-10 negative on Trump coverage.

We have to look at the way the media handled Trump before he was elected. How many newspapers in the entire country endorsed Trump for president? I don’t think the number is zero but it has to be very close to zero.

Some newspapers — The Washington Post and New York Times come to mind — were virulently anti-Trump on the editorial page, which bled over into the news coverage.

Every news network except Fox was anti-Trump, the only positive news being that he was most definitely not going to be elected. Well, actually it was that he was definitely not going to be the Republican nominee, and then that he was definitely not going to be president.

The Huffington Post refused to cover Trump at all as part of its political reporting. All Trump news was published in the entertainment section.

And then he was elected president!

So everyone in the media had to either a) recalibrate their self-image from “We are super smart people who knew in advance that Trump would never be president” to “We are the dumbest people in the universe,” or b) commit to spending the next four years (at least) finding ways to say “we told you so,” e.g. “We told you he was a monster,” “We told you he was a moron,” We told you he was crazy,” etc.

Here’s a chart showing the tone of Trump coverage compared to other recent presidents:

Tone of Trump coverage

The media are setting a new standard for unfavorable press coverage of a president. No president in recent history is even close. Note that the media were mostly cheerleaders for President Obama.


More People I’m Sick Unto Death Of

21 May 2016 /
Redskins primary logo 1972-1981, 1983-present

New poll finds 9 in 10 Native Americans aren’t offended by Redskins nameThe Washington Post

Man, am I sick of people who get offended on behalf of a group they don’t belong to, projecting their own phony outrage on the group members and their own biases on non-group members.

In light of the poll results, non-Native American opponents of the Redskin name seem to have changed their position slightly to say that Native Americans are in fact being offended but are too dumb to realize it.


What Is the Evidence for Unconscious Bias?

21 Feb 2016 /

What is the evidence for “unconscious bias,” since it’s, you know, unconscious?


This Sounds Familiar

12 Aug 2010 /

Bias is particularly an issue on cross-boundary initiatives, where multiple groups of stakeholders work together on a project. Each stakeholder group has particular goals and is judged against metrics that may be unique to that group. Bias tends to be a substantial driver of project disruption in organizations where goals or metrics across groups are in conflict.