Tuesday, March 10, 2020

The Trump Administration And the Coronavirus Pandemic

Two years ago Donald Trump, Our Leader Extraordinaire:

ordered the shutdown of the White House National Security Council's entire global health security unit. NBC News had a good report on this recently, noting that the president's decision "to downsize the White House national security staff -- and eliminate jobs addressing global pandemics -- is likely to hamper the U.S. government's response to the coronavirus."

And how does he now justify that decision?  Like this:

"I just think this is something, Peter, that you can never really think is going to happen. You know, who -- I've heard all about, 'This could be...' -- you know, 'This could be a big deal,' from before it happened. You know, this -- something like this could happen.... Who would have thought? Look, how long ago is it? Six, seven, eight weeks ago -- who would have thought we would even be having the subject? ... You never really know when something like this is going to strike and what it's going to be."

This is the kind of presidential awareness millions of American voters felt comfortable with..

There's a reason why we want smart, knowledgeable and experienced people running countries and important public organizations.  There's a reason why having years of experience matters.  There's a reason for not engaging one's hind-brain in the selection of leaders.

My anger about Trump's incompetence, denials and clear lack of overall understanding comes from seeing how terribly poorly the US is coping, compared to other industrialized countries.  Testing for the coronavirus in this country has been almost a farce so far:

A week later, the United States declared a public health emergency, a process designed to speed the development of diagnostic tests and other medical products. The CDC received the first “emergency use authorization” to make and distribute its test to the backbone of the public health system in the United States — mostly state labs.

But the emergency policy, intended to keep quality high, also discouraged hospital labs from quickly developing in-house tests. They would need specific approval from the FDA to do so.

“Since CDC and FDA haven’t authorized public health or hospital labs to run the tests, right now #CDC is the only place that can. So, screening has to be rationed,” Gottlieb tweeted on Feb. 2.

The CDC manufactured kits, and on Feb. 6 and 7, 90 test kits were shipped to the public health labs. Some labs began to have trouble with the test. On Feb. 12, the CDC announced the test was providing inconclusive results in some laboratories. The problem was in one of the three components of the test.
How do other countries manage to test thousands of people, then?  What about using the test kits they use?  The answer:

Some critics have questioned why the CDC didn’t switch to tests being used by other countries as soon as the problems arose, but the official said it would have taken longer to apply for a new authorization from the FDA and validate and manufacture a new test than it would to fix a test they knew worked in their own lab.

So it goes.  Currently the US testing is too narrow to allow us to tell much anything about the possible spread of coronavirus in different areas.  This means that we have no idea how correct the numbers of those infected might be.  And then there is this bit:

On Feb. 13, HHS Secretary Alex Azar testified before Congress that a limited five-city pilot would begin to add coronavirus to the usual flu surveillance system to see whether “there is broader spread than we have been able to detect so far.” But the plan was delayed because coronavirus tests weren’t available.

Wider testing for the virus is required to properly plan the need for hospital facilities and medical staff and for deciding when stronger quarantine policies might be in order.  So far the US testing has been totally inadequate for these purposes.  To put it into perspective:

As of Sunday, 1,707 Americans had been tested for the novel coronavirus, according to the Centers for Disease Control and Prevention. South Korea, by contrast, has tested more than 189,000 people. The two countries announced their first coronavirus cases on the same day.
This looks like a clown show, to be honest.


Sunday, March 08, 2020

On Gender Norms And Women's Roles. Results From UNDP Surveys.

The New Sex/Gender Norms And Roles Survey:  Main Results

The UN Development Program (UNDP) has published new near-global survey results on social norms about gender.  The findings come from data collected in two waves in the World Values Survey.  The latest wave, from 2010 to 2014, covers surveys done in seventy-five countries.  For a sub-group of thirty-one countries, data is available for both waves in the World Values Survey (2005-2009 and 2010-2014). The latter allows us to see what might be happening over time in various countries about how people, both men and women, view women's proper roles and women's abilities and capabilities.

I spent some time with the results, what with today being the International Women's Day.  How gender roles, norms and gender stereotypes actually work in practice seems a useful topic for thought today, right?  After all, they are one of the main channels which guarantee that women don't stray outside the narrow and rigid traditional sex roles or retrogressive views about what femininity means and how it is properly demonstrated*.

The survey questions try to measure how people, both men and women, view women's empowerment in politics, education, economic life and in physical integrity within a particular culture.  The table below (click to make it larger) shows the questions that were asked:

Both the BBC and the UK Guardian reported on the main findings from the surveys.  Here's the Guardian:

Almost 90% of people are biased against women, according to a new index that highlights the “shocking” extent of the global backlash towards gender equality.
Despite progress in closing the equality gap, 91% of men and 86% of women hold at least one bias against women in relation to politics, economics, education, violence or reproductive rights.

And the BBC:

A new UN report has found at least 90% of men and women hold some sort of bias against females.
The "Gender Social Norms" index analysed biases in areas such as politics and education in 75 countries.
Globally, close to 50% of men said they had more right to a job than women. Almost a third of respondents thought it was acceptable for men to hit their partners.
There are no countries in the world with gender equality, the study found.
The indices referred to in the above quotes are two.  The first one simply counts the percentage of people who hold at least one bias, and the second one counts the percentage of people who hold at least two biases**.

The report itself also summarizes more of the overall findings and also results about change over time in the sub-group of thirty-one countries with data from both waves.  Here are some extra overall findings:

About 50 percent of men and women interviewed across 75 countries say they think men make better political leaders than women, while more than 40 percent felt that men made better business executives. Almost 30 percent of people agree it is justifiable for a man to beat his partner.
Women are skewed towards less bias against gender equality and women’s empowerment. Men are concentrated in the middle of the distribution, with 52 percent having two to four gender social norm biases. More than 50 percent of women are biased in the political arena. Men present biases higher than 63 percent in both the political and economic dimensions, especially for the indicators “Men make better political leaders than women do” and “Men should have more right to a job than women.”Globally close to 50 percent of men agree men should have more right to a job than women.
And here are a few findings about changing social gender norms over time for the thirty-one countries with data from both waves:

More worrying, despite decades of progress in advancing women’s rights, bias against gender equality is increasing in some countries, with evidence of a backlash in attitudes among both men and women.
According to the GSNI2, the proportion of people with moderate and intense biases against gender equality grew over the last few years in 15 countries (out of 31). The share of both women and men worldwide with moderate to intense gender biases grew from 57 percent to 60 percent for women and from 70 percent to 71 percent for men (table 3). Surveys have shown that younger men may be even less committed to equality than their elders.34

Progress in the share of men with no gender social norms bias was largest in Chile, Australia, the United States and the Netherlands (figure 7). At the other extreme, indicating a backlash, the share of men with no bias fell in Sweden, Germany, India and Mexico. The share of women with no gender social norms bias increased the most in the Netherlands, Chile and Australia. But most countries in the sample showed a backlash, led by Sweden, India, South Africa and Romania (see figure 7).

All the above findings matter, of course.  But it's worth diving much deeper into these survey results, to see stuff that becomes invisible in the averaging process that produced those frightening overall percentages.   That's what I do in the next section.

Individual Country Data And Change Over Time

Certain overall patterns are visible across all countries in both the waves of the World Values Survey.  The most important one is that women, almost everywhere,  are less likely to hold the biases the report analyzes than men, though the differences are not large.  That the male and female average biases go together within each country demonstrates the importance of culture, tradition and religion.  All of us grow up absorbing the cultural and religious norms of our communities, after all.

To give an extreme example of this, consider the respondents without any gender social norm bias in the country data of the surveys.  Sweden, a gender-egalitarian country, has extremely high percentages of both men and women who expressed no gender bias in the two waves of the survey, though those numbers declined from the first two wave to the second: 

In 2004-2009 82.2% of women and 79.79% of men showed zero gender bias, while in 2010-2014 the respective percentages fell to 71.69% and 68.29%.

In contrast, corresponding numbers for Jordan, a country with little gender-equality, looked very different.  In 2004-2009 the percentage of both men and women without any gender norm bias was 0.4%, and in 2010-2014 rather similar results applied, with 0.83% of women and 0.5% of men showing zero bias.

The point I wish to make is that the overall averages reported in the BBC and the Guardian hide large amounts of variation between countries

Some countries (Jordan, Ghana, India, Malaysia, Rwanda, South Africa and Turkey among the countries for which data is available for two waves***) show very strong bias against women's empowerment (so that the percentage of respondents showing no bias at all is at most only a few percentage points), while some other countries (Sweden, Australia and the Netherlands among the countries for which data is available for two waves) show strong social agreement about the desirability of social, political and economic equality between men and women (so that the percentage of either men or women or both who hold no bias exceeds fifty percent).

Most countries fall between these two extremes.  The United States comes closer to the latter group than the former, with 46.09 % of women and 39.08% of men showing no bias in the 2010-2014 wave of the survey.

The United States is also one of the countries for which reported bias shrank between the two survey waves, for both men and women.  The change was especially large for men (from 33.06% to 39.08%).  This is, of course, great news.

The following picture demonstrates changes in how respondents view gender norms about women between the two survey waves:

In interpreting it, keep in mind that, for instance, Swedish men and women still hold much more gender-equal views than Chilean men and women.
In general, I would counsel caution in how we would interpret the data on changing gender norm bias over time.  Stuff like this, reproduced from a quote above:
Progress in the share of men with no gender social norms bias was largest in Chile, Australia, the United States and the Netherlands (figure 7). At the other extreme, indicating a backlash, the share of men with no bias fell in Sweden, Germany, India and Mexico.
That's because this particular survey has only been done twice.  Specific events (and news about them) immediately preceding a particular wave of the survey could influence the answers of the respondents in that wave, even if there was  no longer term trend toward more or less bias against women's empowerment.  It's also worth checking if demographic changes in the populations to be surveyed are controlled for before interpreting the findings as changes in the gender norm biases rather than in the population composition.

Relating to all this, note that the general dismal findings might perhaps not be so dismal.  This, too, has to do with the fact that these surveys are very recent.  My guess is that data from earlier times would have shown even more gender norm bias.

On Descriptive And Normative Bias

What does all this data really mean?  It's important to note that the questions respondents were asked could elicit both descriptive and normative bias.  The former relates to how things actually are in a particular society, the latter to how things "ought to be" in that same society.

It is possible that some respondents interpreted in the results as holding biases are answering on the basis of descriptive bias.  As an example, if you live in a country where men are almost all the bread-winners stating that it's more important for men than women to get a university education might just be a statement of fact, not an expression of one's normative beliefs of how that society should be arranged or how women's lives should be regulated.  Likewise, a respondent agreeing that men make better political leaders than women gives a statement hard to interpret if no woman has ever held political power in that respondent's country.  How does one evaluate a statement like that without any data?  Perhaps by assuming that the way things are is based on actual male advantage?  But that assumption hasn't actually been tested.

Final Comments

I hope you read this far.  I almost didn't write this far, what with my body falling apart and so on.  The take-home message to me, after wading through lots of tables and text, is not to take it for granted that equality between men and women will just go on increasing, rather than stalling or even reversing.  In fact, the survey itself notes that advances appear to have slowed down and may have even ended in some parts of this earth.

For those wishing a happier concluding comment, you can always look at the complement of the overall findings:  If one third of the respondents seem to think that men can use physical violence against their partners, then two thirds don't think that way, and if almost half of all men think that men have more right to a job than women, then more than half of all men don't think that way.

*  In my view getting rid of all the gender roles, norms and stereotypes about femininity and masculinity that we possibly can is the only functional road to a more feminist and fairer world.

** The cutoff points and aggregation methods used in the report when constructing the two indices from qualitative answers are, of course, to some extent based on choice. Some choices are easier than others.  For instance, to call an answer biased when it "strongly agrees" or "agrees" that men's access to jobs matters more is a fairly straightforward choice.  But what does one do when someone's agreement or disagreement with an assertion is a number picked from a ten point scale?  Where, on that scale does one draw the bias line?

***  This choice is based on nothing but my current lassitude, caused by ill health.  All the same stuff can be seen in the larger data set, but I wasn't going to go and copy numbers twice, sorry.  Rests weary head against the cool pillow.