Analytics in Government

[WeezyBengal]

With sports and so many other facets of every day life relying on data and analytics to makes key decisions...when will our government and a President get on board with a fully analytical approach to policies? I'm not sure how much and what is being done at this moment with big data in government...but could you imagine if a President ran the country like some of the new age GMs in baseball ran their teams?

It just seems like so many industries nowadays take opinion and bias (pretty much what politics is) out of the equation and rely solely on factual data to make key decisions and to find REAL answers.

If a President ran independently and was committed to finding answers to our problems through analysis, data, and factual information - basically letting the data and analytics decide on which policies are best for the American people - would you vote for them?

[fredtoast]

Nothing could be a more polar opposite then the situation we have now where elections are about a cult of personality salesmanship instead of policy based.

Voters don't even care if a candidate is telling the truth as long as they like "the way he says it".

[WeezyBengal]

(06-29-2018, 11:22 AM)fredtoast Wrote: Nothing could be a more polar opposite then the situation we have now where elections are about a cult of personality salesmanship instead of policy based.

Voters don't even care if a candidate is telling the truth as long as they like "the way he says it".

I agree with that. 

I also think it would be interesting to see what would happen if someone REALLY pushed this approach while they were campaigning. If they really pushed decision making based on analytics and using an analytics team. That''s never really been done before. So...it could be a type of salesmanship and people might get behind it. You are basically selling people on taking bias and opinion out of policy making and creating policy for people that works based on factual information. So, it kind of is a selling point... 

[Belsnickel]

Lemme get ready, here.



So this is, essentially, what I study. Public policy analysis/assessment is all about using data to determine the best possible policy solutions for implementation as well as evaluating current policy that is in place so as to improve outcomes related to effectiveness, efficiency, and equity. This is something used a lot in our government below the levels of the elected officials. Career staffers in particular utilize these tools to help them adjust the implementation of laws. There are even agencies within the government devoted to this sort of thing. The Congressional Research Service and the Government Accountability Office are just two stand-alones that report to Congress with two different roles, but most agencies also have wonks within their ranks that do in house work for them.

Here is the problem with what you suggest. A government run in this manner is called a technocracy (see my title to the left). It is a government run by experts using the data they come up with. This is all fine and dandy on the surface, seems like an okay thing. But our government is not a technocracy, it is a democracy. A democracy demands that the people be the ultimate determination of the actions of our government. In a technocracy, the data and the experts rule and what the people want doesn't matter. That's not our government, and it also isn't as effective as we'd like to think. I know, that seems weird coming from someone that self identifies as a technocrat.

Technocrats are important tools in a democratic government. Like what the CRS does, we provide information for elected officials to help make their decisions, but they must also be guided by the agendas in play. These agendas are the ones guided by the parties, but also by the people and society as a whole. There is only so much time they can spend on the floor of their legislative body (though more if they weren't busy taking bribes seeking donations) and so these agendas determine what things they will focus on. The current mood may also indicate that some optimum solutions may not be acceptable to the public at this time and so that legislation is pocketed for a time down the road when maybe they can sway public opinion, more.

This, of course, is all in theory. This is how our government should work, and would work were it not for the influence of corporations and the wealthy. Currently, wonks in the government are being ignored and lobbyists are replacing them. Instead of relying on technocrats, experts in their field in the government, to help craft good policy our elected officials are allowing special interests to do this for them. What is happening is that wonks cut their teeth in government and then jump ship for a much larger paycheck with a lobbyist. They utilize their access, relationships, and acquired knowledge to seek out the elected officials most favorable to their cause and hand them biased information so that they get a win for their employer and a nice fat paycheck.

It should be noted that FDR did this sort of thing. This is what his "Brain Trust" was all about. These days, instead of experts as political appointees we get big money donors with ties to Wall Street. And that is a knock against Clinton and Obama as much as anyone else.

[fredtoast]

Look what happened when they tried to use science to address climate change.

Many people still swear that for profit oil companies are the altruist and that government public servants are "in it for the money".

People don't care about facts.  They care about the beliefs of their tribe.

[Belsnickel]

This thread was a great idea, BTW. Sorry if I kind of killed it with my wall of text answer.

[fredtoast]

(06-29-2018, 12:12 PM)Belsnickel Wrote: This thread was a great idea, BTW. Sorry if I kind of killed it with my wall of text answer.

The problem was not the size of your post*.  It was the fact that it did not include either "Trump is a Nazi!!" or "What about Obama!!!"









* That's what she said

[*TheLeonardLeap*]

Analytics would 100% be seen as a way for "racists" to run the government.

"Sorry African Americans, but your High School graduation rate is the lowest on the team, and your murder rate is absurdly high. We're going to have to cut you to make room for more Asian Americans. Work on making more two-parent families in the offseason and maybe we'll call you in for a workout to see about getting you a non-roster invite to Spring Training."


- - - - -

It really only works in sports because we expect teams to be relatively brutal about their management in an effort to win above all else.

"Analytics say your hard-hit % and your O-Swing % are in the bottom tenth of hitters, so we don't think your BABIP is low due to bad luck, but more a product of your hitting profile. We're sending you down to AAA until you hit the weights harder and work on better strikezone recognition."

...and then they never get called up again, 2 years later get traded to another team for a PTBNL, and fade into obscurity a year or two after that, before pursuing a job selling cars back in their hometown. We see that and think "yes, that is how it should be, because we need to win". It wouldn't really work for Government.

[*michaelsean*]

(06-29-2018, 12:12 PM)Belsnickel Wrote: This thread was a great idea, BTW. Sorry if I kind of killed it with my wall of text answer.

Are you kidding?  When I saw the topic, I was prepared to get hit with 50,000 words.  

[fredtoast]

(06-29-2018, 01:49 PM)TheLeonardLeap Wrote: Analytics would 100% be seen as a way for "racists" to run the government.

"Sorry African Americans, but your High School graduation rate is the lowest on the team, and your murder rate is absurdly high. We're going to have to cut you to make room for more Asian Americans. Work on making more two-parent families in the offseason and maybe we'll call you in for a workout to see about getting you a non-roster invite to Spring Training."

I am thinking that it would be a chance for the government to address some of the reasons behind low test scores for African Americans.

Analytics would address underfunded school systems and other socioeconomic issues instead of saying "blacks are just an inferior race".

But there would be many unpopular decisions.  For example men are the most violent and criminal demographic group.  The difference between men and women in violent crime rate is much larger than the difference between any two races.  So a lot of white guys who like to point at the issue of high crime races would not be pleased with steps taken to address the crime rate of their sex.

[*TheLeonardLeap*]

(06-29-2018, 02:27 PM)fredtoast Wrote: I am thinking that it would be a chance for the government to address some of the reasons behind low test scores for African Americans.

Analytics would address underfunded school systems and other socioeconomic issues instead of saying "blacks are just an inferior race".

But there would be many unpopular decisions.  For example men are the most violent and criminal demographic group.  The difference between men and women in violent crime rate is much larger than the difference between any two races.  So a lot of white guys who like to point at the issue of high crime races would not be pleased with steps taken to address the crime rate of their sex.

Except in a sports analytic world, they will tell you what you need to work on, and even give you some help, but it's on you to put in the work to make yourself improve. (See Dalton going to that guy in the offseason to improve his deep throws).


Yeah, men being the source of a large majority of violent crime is another reason why analytics simply wouldn't work for running a government.

Or the abundance of HIV in the gay community.

Or the abundance of drug use in poor communities. (Illegal Substance suspensions?)



- - - - -

Either way you slice it, running a government based off of stats like a highly analytical sports team probably doesn't really work and there would likely be very few people who would be happy in the end... except maybe straight upper-middle-class Asian American women. Lol

[Belsnickel]

(06-29-2018, 01:49 PM)TheLeonardLeap Wrote: Analytics would 100% be seen as a way for "racists" to run the government.

"Sorry African Americans, but your High School graduation rate is the lowest on the team, and your murder rate is absurdly high. We're going to have to cut you to make room for more Asian Americans. Work on making more two-parent families in the offseason and maybe we'll call you in for a workout to see about getting you a non-roster invite to Spring Training."


- - - - -

It really only works in sports because we expect teams to be relatively brutal about their management in an effort to win above all else.

"Analytics say your hard-hit % and your O-Swing % are in the bottom tenth of hitters, so we don't think your BABIP is low due to bad luck, but more a product of your hitting profile. We're sending you down to AAA until you hit the weights harder and work on better strikezone recognition."

...and then they never get called up again, 2 years later get traded to another team for a PTBNL, and fade into obscurity a year or two after that, before pursuing a job selling cars back in their hometown. We see that and think "yes, that is how it should be, because we need to win". It wouldn't really work for Government.

This works on the assumption that we would not take these sorts of things into account. Public policy is a social science, and we incorporate psychology, sociology, economics, etc., into our research. It's one reason I enjoy it because you get to dabble in a bit of everything. This school system that is predominantly black is underperforming? Why? Does it have to do with family situations, single family households, lack of a living wage for parents?

Technocratic solutions aim to get at the root of the problem much more than democratic solutions do and our ethical standards require working towards equity. I'm not saying it's perfect, but there is less of a chance of racist policy in that situation than with elected officials. The reason the SCOTUS is there and the reason for the Senate are both to protect the minority from the majority.

[*TheLeonardLeap*]

(06-29-2018, 03:28 PM)Belsnickel Wrote: This works on the assumption that we would not take these sorts of things into account. Public policy is a social science, and we incorporate psychology, sociology, economics, etc., into our research. It's one reason I enjoy it because you get to dabble in a bit of everything. This school system that is predominantly black is underperforming? Why? Does it have to do with family situations, single family households, lack of a living wage for parents?

Technocratic solutions aim to get at the root of the problem much more than democratic solutions do and our ethical standards require working towards equity. I'm not saying it's perfect, but there is less of a chance of racist policy in that situation than with elected officials. The reason the SCOTUS is there and the reason for the Senate are both to protect the minority from the majority.

I am almost certain they already do studies like that and have all kinds of stats on things like that.

I thought you were talking what it would be like literally running the country like an analytic heavy sports team. Lol

[fredtoast]

(06-29-2018, 03:17 PM)TheLeonardLeap Wrote: Except in a sports analytic world, they will tell you what you need to work on, and even give you some help, but it's on you to put in the work to make yourself improve. (See Dalton going to that guy in the offseason to improve his deep throws).

I don't see any difference between teams providing weight rooms, training rooms, rehab facilities, film rooms, game films, training table meals, doctors, psychologist, computer motion analysis and other resources to the government providing resources to schools, prisons, and drug rehab facilities.  In both cases the individuals need resources, but they also nee to work to succeed.

[WeezyBengal]

(06-29-2018, 12:12 PM)Belsnickel Wrote: This thread was a great idea, BTW. Sorry if I kind of killed it with my wall of text answer.

LOL. No, it makes sense. Never heard of the Techocrat thing, but I get where you are coming from and it is interesting. You bring up valid points as to why it wouldnt work. Hopefully there can be a good mix of democracy and "techocracy" in the future. 

[*michaelsean*]

(06-29-2018, 04:10 PM)WeezyBengal Wrote: LOL. No, it makes sense. Never heard of the Techocrat thing, but I get where you are coming from and it is interesting. You bring up valid points as to why it wouldnt work. Hopefully there can be a good mix of democracy and "techocracy" in the future. 

Technocrats are the people bureaucrats listen to when they can’t get to sleep.

[Belsnickel]

(06-29-2018, 04:03 PM)TheLeonardLeap Wrote: I am almost certain they already do studies like that and have all kinds of stats on things like that.

I thought you were talking what it would be like literally running the country like an analytic heavy sports team. Lol

Yeah, we do have all sorts of stats on things like that. I think I look at this a bit differently than most because I am the type of person that would be doing these analytics and writing policies with them, so I have a deeper working knowledge of how it all works.

(06-29-2018, 04:10 PM)WeezyBengal Wrote: LOL. No, it makes sense. Never heard of the Techocrat thing, but I get where you are coming from and it is interesting. You bring up valid points as to why it wouldnt work. Hopefully there can be a good mix of democracy and "techocracy" in the future. 

That is the ideal. What we strive for are evidence based policy solutions to items on the legislative agenda.

So elected officials care about an issue because they are being yelled at by mouthy jagoffs contacted by concerned citizens. We research the issue and run statistical modeling and come up with (potentially) several options that would be viable. It is then up to the legislature to craft the policy using this information and debate and come to compromises that happen in a healthy, functioning democracy. The solutions are evidence based, the information is coming from career staffers that are experts in the field and also in the implementation of the type of policy. There is much rejoicing because things are working as they should.

Alternatively, there are think tanks that can also help in the wonkiness. Brookings, for instance, was created to provide a source external to the government to provide unbiased evidence based policy decisions. Since then, other think tanks have cropped up. Some good, some not as good. One easy way to tell if a think tank is reputable is to look online for their financials. If they spend more money on communications/education (i.e. marketing their ideas) than on the actual research (coughcoughHeritageFoundationcoughcough) that is a good sign that they aren't necessarily focused on working towards evidence based solutions.

[Dill]

(06-29-2018, 01:49 PM)TheLeonardLeap Wrote: Analytics would 100% be seen as a way for "racists" to run the government.

"Sorry African Americans, but your High School graduation rate is the lowest on the team, and your murder rate is absurdly high. We're going to have to cut you to make room for more Asian Americans. Work on making more two-parent families in the offseason and maybe we'll call you in for a workout to see about getting you a non-roster invite to Spring Training."


- - - - -

It really only works in sports because we expect teams to be relatively brutal about their management in an effort to win above all else.

"Analytics say your hard-hit % and your O-Swing % are in the bottom tenth of hitters, so we don't think your BABIP is low due to bad luck, but more a product of your hitting profile. We're sending you down to AAA until you hit the weights harder and work on better strikezone recognition."

...and then they never get called up again, 2 years later get traded to another team for a PTBNL, and fade into obscurity a year or two after that, before pursuing a job selling cars back in their hometown. We see that and think "yes, that is how it should be, because we need to win". It wouldn't really work for Government.

Excellent analogy. Excellent point.

[Dill]

(06-29-2018, 11:18 AM)WeezyBengal Wrote: With sports and so many other facets of every day life relying on data and analytics to makes key decisions...when will our government and a President get on board with a fully analytical approach to policies? I'm not sure how much and what is being done at this moment with big data in government...but could you imagine if a President ran the country like some of the new age GMs in baseball ran their teams?

It just seems like so many industries nowadays take opinion and bias (pretty much what politics is) out of the equation and rely solely on factual data to make key decisions and to find REAL answers.

If a President ran independently and was committed to finding answers to our problems through analysis, data, and factual information - basically letting the data and analytics decide on which policies are best for the American people - would you vote for them?

No. 

How might some groups react when "analytics" decides they need to pay more taxes to support public education, or that their wages need to be lowered, or that their health care should not cover birth control, or that women should be paid more than men for the same work, or that people over 60 have to retire, or that more women, or less, should have abortions?


This proposal seems to me to confuse judgments of fact with judgments of value, though these are wholly separate orders. And a pluralist society like ours is made up of differing demographics, differently empowered, and embracing sometimes very different values, which means group interests must sometimes (or of continuously) come into conflict.  Machines cannot make value judgments, even if they can sometimes be programmed to approximate them when parameters of judgment are simple and agreed to by all parties, as in Leonard's sports example.

[*TheLeonardLeap*]

(06-29-2018, 10:51 PM)Dill Wrote: No. 

How might some groups react when "analytics" decides they need to pay more taxes to support public education, or that their wages need to be lowered, or that their health care should not cover birth control, or that women should be paid more than men for the same work, or that people over 60 have to retire, or that more women, or less, should have abortions?


This proposal seems to me to confuse judgments of fact with judgments of value, though these are wholly separate orders. And a pluralist society like ours is made up of differing demographics, differently empowered, and embracing sometimes very different values, which means group interests must sometimes (or of continuously) come into conflict.  Machines cannot make value judgments, even if they can sometimes be programmed to approximate them when parameters of judgment are simple and agreed to by all parties, as in Leonard's sports example.

I know a machine that values your patience while it works on wiping out the human race.