The Individual and the Bathwater 2019-11-29T14:17:32.093Z · score: 50 (25 votes)


Comment by tom_beggs on Correlations Between Cause Prioritization and the Big Five Personality Traits · 2020-10-22T20:10:16.706Z · score: 4 (3 votes) · EA · GW

No problem, best of luck! :)

Comment by tom_beggs on The Intellectual and Moral Decline in Academic Research · 2020-10-01T18:11:52.982Z · score: 2 (2 votes) · EA · GW

I wouldn't expect a marked difference in the quality of non-controversial research whether funded by a national granting agency or private industry. That said, I'm not an expert on the topic, either.

As for "controversial" science in the sense of "any science that business/industry doesn't like," the pattern is quite similar whether we're talking about lead, asbestos, climate change, et cetera:

Find a couple of researchers who will play ball to say your product is safe despite all the evidence to the contrary. Point to this repeatedly any time the topic comes up. Ignore the mountain of evidence that says you're wrong at all costs, and undermine it any way you can. Buy as many politicians as you can to try and prevent regulation.

As for vaping, I'd need to see some examples of bad academic research. I'm not sure you can blame the scientists for the consequences of poorly-regulated businesses. They can only test the products that they're given and tell you whether they're safe. They can't tell you if a manufacturer is going to change the oil they use or decide to include heavy metals.

That's up to the regulators to prevent.

Comment by tom_beggs on The Intellectual and Moral Decline in Academic Research · 2020-09-28T15:03:03.800Z · score: 1 (1 votes) · EA · GW

Without a Google Scholar search, I'd just point to how industry dealt with things like asbestos, lead, tobacco, and greenhouse gas emissions.

Science definitely needs scrutiny, debate, and evidence. That said, whenever someone's loudly proclaiming that an entire field is corrupt and incorrect, it should raise suspicions.

What's more likely: that thousands of (generally underpaid) researchers in hundreds of competing labs worldwide are all in cahoots, or that a few dissenting voices are funded by industry to argue that point of view?

Comment by tom_beggs on Correlations Between Cause Prioritization and the Big Five Personality Traits · 2020-09-25T19:35:41.761Z · score: 6 (4 votes) · EA · GW

Hi David,

I agree that pitting the "top priority" group against the "no resources" group in a correlation is an odd way of approaching the data. I was trained not to create dichotomous variables unless advanced statistics tell us that the data is, indeed, dichotomous. Doing so gets rid of variance that may well be useful. The current analysis goes even further by excluding large numbers of people, as you noted.

I would be interested in seeing non-parametric correlations that can deal with ordinal data (e.g, Spearman's) used on the full sample. I'm guessing the story may be different.

Comment by tom_beggs on How have you become more (or less) engaged with EA in the last year? · 2020-09-11T14:02:12.891Z · score: 5 (4 votes) · EA · GW

From your article on the male variability hypothesis:

Recent studies indicate that greater male variability in mathematics persists in the U.S., although the ratio of boys to girls at the top end of the distribution is reversed in some specific immigrant groups.


These results have been replicated and expanded in a 2019 meta-analytical extension [...], which found that policies leading to greater female participation in the workforce tended to increase female variability and, therefore, decrease the variability gap.

These sound like arguments for environment to me, which would mean that "Frank" is still likely correct and "Grace" is misinformed on what the patriarchy is and how it works.

Comment by tom_beggs on The Effects of Animal-Free Food Technology Awareness on Animal Farming Opposition · 2020-05-20T22:08:35.469Z · score: 3 (3 votes) · EA · GW

Thanks for the interesting article, Jamie!

I'd be inclined to think that it's the second explanation offered in the discussion that's driving the effect--naturalness bias happens in a lot of areas, and the AFFT texts were pretty lab/tech-heavy. It would be easy to test whether the first mechanism you proposed is correct by asking people whether they think farming will be replaced with these technologies. Future studies could also present the AFFTs in a less science-heavy way to see if the effect is attenuated. :)

Comment by tom_beggs on The Intellectual and Moral Decline in Academic Research · 2020-02-12T23:40:09.962Z · score: 2 (2 votes) · EA · GW

Yeah, the more I looked into the guy, the more his critique fit into context. His work finds a home on some websites of questionable repute. haha

And as you point out, the people you meet in academia generally don't tend to be as he's characterized them.

I would be willing to bet that he has a financial motive to argue against the prevailing scientific consensus, just as we see in other instances where facts turn out to be inconvenient for corporate interests.

Comment by tom_beggs on The Intellectual and Moral Decline in Academic Research · 2020-02-12T23:05:43.938Z · score: 2 (5 votes) · EA · GW

Thanks for your feedback.

I was trying to figure out why the author would be so, so critical of scientific research.

I would say he was downright uncharitable, in fact.

It turns out that he's also argued quite strongly that high levels of refined sugar in people's diets are no problem: e.g.,

To do so, he has to throw aside mountains of scientific research. I would say his attack above is a necessary part of that effort.

So while I am concerned about inefficiencies in academic work and the waste of taxpayer dollars, I'm much more worried about the effects of corporate money on research.

Comment by tom_beggs on The Intellectual and Moral Decline in Academic Research · 2020-02-10T18:56:53.956Z · score: 3 (3 votes) · EA · GW

I wouldn't say that there are no inefficiencies in academia. There are inefficiencies in every line of work.

I would say that on the whole, a lot of great still work gets done.

I definitely wouldn't say that academia is rife with "incompetence in concert with a lack of accountability."

Sure, there are people with Ph.Ds who are not strong researchers. There are lot of them who are, though.

We may just disagree on the ratio of the two groups based on our own experiences.

Comment by tom_beggs on The Intellectual and Moral Decline in Academic Research · 2020-02-07T19:39:06.189Z · score: 9 (14 votes) · EA · GW

I would argue the article is extremely pessimistic.

Yes, funds sometimes get misallocated or are given to people who have committed fraud.

More often, they go to hard-working researchers who really don't make that much at all...people who hate fake or misleading scientific claims more than the average taxpayer.

And yes, there's a replication crisis...that people are aware of working to address.

In short, I think the author uses an extremely broad brush: "The widespread inability of publicly funded researchers to generate valid, reproducible findings is a testament to the failure of universities to properly train scientists and instill intellectual and methodologic rigor."

And yet, scientific breakthroughs happen all the time and the world is better for it.

In short, maybe the author is burnt out or has only ever worked with poor colleagues? Or hasn't been funded in a while?

Most of the researchers I've met are honest and hard-working and doing their best to get it right, even in the face of challenging questions and strained resources.

Comment by tom_beggs on The Individual and the Bathwater · 2019-12-18T17:07:39.059Z · score: 5 (4 votes) · EA · GW

Hi Jacob,

Thanks for the thoughtful reply! I’ll address each point one at a time. Your comments are in bold:

First, I'm not sure I follow the "Institutions are likely aiming for "good enough."" section: if an improvement in animal welfare is profitable, it should presumably happen without any advocacy. But I'm not sure it then follows that "the pressure of public opinion is needed to drive welfare beyond "good enough".

The last line is really close to the central thesis of the work. In this section, I discuss the fact that agri-corps and governments may be driven in part by concerns about health and antibiotic resistance, which would increase the tractability of institutional campaigns. That boost would be short-lived once those relatively low bars are met.

Most asks we make of corporations will likely not increase profits. In order to minimize any negative financial impact, institutions will do the least amount needed ("good enough") to maintain positive public opinion. Additional public pressure will be needed to take the next step, and so on.

Second, most corporate cage-free commitments don't rely on consumer “willingness to pay” for cage-free eggs per se. Instead of asking people to directly choose cage-free over caged eggs, entire restaurant chains and states provide exclusively cage-free eggs, so the choice would have to be made at the restaurant or state-level. It’s possible people would then choose not to buy certain products or patronize particular restaurants, but given the low price elasticity of eggs I wouldn’t expect this to be a large effect; it seems even less likely that people in Los Angeles or San Francisco would drive hours to obtain modestly cheaper eggs.

I think restaurants could face the same type of pressure that producers do: if you’re putting a more expensive product on the plate, you need consumers to support those costs. If they don’t, your competitors have an advantage. The choice of a slightly more expensive breakfast that features more ethical eggs is still a willingness to pay. (That said, the eggs are only one of the things on the plate, so this effect is diluted. It’s also entirely possible that cage-free eggs would attract as many--if not more--diners as they discourage, although I’m not aware of any research on that addresses that question. All-told, I think that restaurants may be a bit more immune to the proposed effects than grocery stores or whole markets. Also, eggs may not be the best example for restaurants given the small impact on consumers, but future welfare reforms that affect meat may have more impact on costs and therefore consumer reactions.)

Now, passing state legislation is obviously wonderful. And no, I really don’t think people will drive for hours to buy cheaper eggs. The point I was making was a general one about holdouts, rather than a specific argument relating to Prop 12.

That said, it’s worth noting that Prop 2, the precursor to Prop 12, originally only covered Californian production. This meant that non-compliant eggs produced in other states would still have been sold in California. Therefore, the mechanism I proposed was in place (until a new law was passed to correct the issue). My understanding of Prop 204 in Arizona suggests it is subject to the same limitations.

Prop 12 also represents a “willingness to pay” among the Californian public in my opinion because they were already well aware of the cost premium of cage-free eggs based on their experiences with Prop 2. Specifically, Prop 2, which had fewer protections, had already led to a 22% average increase in egg prices. This increase would have been taken into account by most voters when voting for the more expansive measure. There were also industry efforts to highlight the additional increases of Prop 12 to the public.

I think this interpretation that corporate campaigns wins have been caused by the attitudes, intentions and behaviors of the public is likely where we disagree most. I agree there is likely some requisite threshold of public support, but it’s not clear this public support then causes change.

I don’t think I implied that individuals caused institutional wins like Prop 12, but feel free to correct me on that. The legislation originated in the Humane Society’s Prop 2 as I understand it.

However, Prop 12 only passed because individual people cared enough to vote for it. In short, one of the biggest institutional wins of them all so far was completely reliant on voters.

If other states have similarly high levels of support for such an initiative and the political will to put it on a ballot, then that’s great, and something we should definitely pursue if the change for animals is meaningful enough to warrant the investment! In the states that don’t meet these requirements, we have more work to do. That’s where we still need to be shifting the individual.

(And we also have to consider whether the current ballot initiatives meet our end-goal, and if not, whether we currently have the necessary public support to get there. I would argue again that we likely still have work to do. While the numbers you shared look good in the abstract, I think there’s a lot of work needed to convert that support into meaningful legislation.)

Third, I agree with the concern about holdouts. However, this is where the power of legislation, and, unless that holdout is a producer, secondary targeting can come in to play.

Fair enough, and it would depend on the proportion of holdouts for a given reform.

Fourth, at least where the industry has actually transitioned to cage-free (about 20% currently), I’m less concerned about “institutional backslide” as this would again require large capital investments to reinstall cages.

I agree that no corporation would re-renovate (unless it made financial sense to do so). I am worried about stopping current renovations and scrapping plans for future improvements, as we have seen happening (e.g., Bloomberg, 2017, as cited above).

Also, some of the “commitments” of some of the very largest organizations like Walmart have left themselves a lot of wiggle room to back out if price or demand aren’t there to support the change. The language makes that explicit: “Transition to a 100 percent cage-free egg supply chain, based on available supply, affordability and customer demand by 2025.” They could walk away quite easily if any of the three conditions were not met.

ACE has estimated that just over 50% of organizations will keep their commitments, and no one can estimate how lack of follow-through on this campaign will affect future campaigns, so we are concerned about downstream effects of this type of backsliding. Even Prop 12 is going to be challenged in the courts.

Fifth, I think we should not take the industry's claims (like “In response to initially weak demand for cage-free eggs...”) at face value, and there are many possible reasons a particular producer might not remove cages. (The article offers some other explanations, like the wide price spread resulting from the post-avian influenza egg glut. But we should also consider more mundane possibilities like Rose Acres creditors weren't willing to lend them more money—it certainly looks better to say there was insufficient consumer demand!)

That certainly could be the case, but creditors generally have no problem shelling out for something that is either lucrative or that leads to greater market share.

As for avian influenza, as this industry report shows, the price gap between conventional and cage-free eggs decreased because of the avian flu crisis, but then increased again as flocks recovered.

Lastly, I'm not sure which particular outcomes you're referring to with gestation crate commitments, but the details of the pork industry and the commitments create some dissimilarities with the cage-free commitments.

Same as with eggs: increased costs and issues around supply and demand.

I don't think anyone is suggesting shifting "most of the resources" (Jacy only suggests 50% of existing individual resources) and certainly not all resources, so I don’t think the "and" not "or" message is really relevant. Of course, if there are people who hold this view I’d be curious to learn more. I think the question for most is identifying the optimal ratio. Sentience Institute is also quite clear on the evidence underlying that belief and it's not a lack of a link between advocacy and diet, but the existence of more public support for institutional change, historical precedent and psychological arguments. For what it’s worth I also believe we need some balance between the two, as Harish Sethu wrote about at Labs.

To be precise on the language, Reese (2018) says “We need to shift over 50% of the resources and messaging we currently use to promote individual change.” When giving examples of the kinds of dietary change efforts he recommends, he says “On the final page of a leaflet, for example, it may be best to offer several concrete calls to action, one of which can be individual diet change.” So individual change (which is defined as dietary change) is now a point on the back of a leaflet.

(Despite this narrow definition of individual work, some articles take it to mean that individual approaches are likely to be less effective than institutional approaches. I think part of this is confusion around the conflation of one type of behavior change with all individual efforts. As Jamie and I discovered on the other post, Sentience’s definition of individual work is quite narrow and mine is quite broad. That definition is perhaps where the main disagreements between us lie as well.)

Also, the psychological evidence provided for the claim in the Reese post is quite broad: one could choose other psychological theories and evidence to generalize and make similar arguments in the opposite direction. In short, we need to use better evidence to make our claims, and if such evidence doesn’t exist, we need to be creating it.

I’m also skeptical of the arguments based on historical social movements. Simply put, while social movements don’t generally focus on behavioral change, they are still reliant on it. Martin Luther King Jr. would not have gotten far if people hadn’t stood up to march with him. Yes, the moment of victory comes in the form of legislation. But without individuals, that day would simply not have arrived.

I’m most convinced by the argument around existing public attitudes, which I’ll address in response to your comment below.

All that said, I would still consider the leaflet that was given as the example in the article as an individual outreach effort: it looks to create support for ballot initiatives and to create new advocates. And I agree that diet change is likely the hardest, final step to reach. As I argue above, we still very likely need to reach it in many cases.

So all-told, you may be right that the either/or message may not be accurate. I liked this line from the HLL link you provided: ”When outcomes of interventions are interdependent, the effectiveness of each is inextricably linked with those of the others. Justifying one as being more effective than another is not quite straightforward — declaring so is often misleading.” I think I was getting the impression that some were starting to make declarations. I also wholeheartedly agree with this line: “Not only is the likelihood of the success of each form of advocacy increased by the success of the other, each of them is crucial to the other for sustained progress toward their shared goal.” The last half of the quote is what I was trying to argue above.

It seems that quantifying the relative balance may help alter the misperception that can arise from more general arguments. With that in mind, it might be helpful for the discussion if you could define what you consider to be “individual” approaches and “institutional” approaches and suggest roughly how much of the resource pool you would assign to each. Your arguments here all seem to point in one direction, so I’m curious where the balance lies for you--although you may have perceived me as trying to pull the pendulum in the other direction (when I was more trying to stop it in the middle) and been presenting an argument from that perspective. We may be closer to agreement than it appears.

I appreciate your pointing out that many institutional approaches have “even less real-world empirical support.” As you know, The Humane League Labs is actively working to rectify this :) However, one might object that in the history of social change, institutional approaches have significantly more empirical support, at least in so far as few if any social movements seem to have succeeded by convincing each individual to their cause. I think there is still work to be done marshalling this evidence and I’m not convinced the existing evidence is applicable to animal advocacy. But I nonetheless find this evidence (the history of social change) more convincing than not in support of institutional approaches (maybe 60% credence).

I don’t think that social movements need to convince each individual. I don’t think they realistically can. I do think we need to convince enough people, though.

As above, neither MLK Jr. or Prop 12 would have gotten anywhere without the requisite support. And while sometimes that already exists and just needs to be tapped, other times it needs to be created. (It’s also worth noting that the reason this support exists is in no small part due to years of effort.)

I am also concerned that it’s not clear what individual efforts you’re suggesting

I guess in general, we need to be increasing our knowledge of people’s attitudes, beliefs, feelings, intentions, and behaviors (including donations, signing petitions, voting, activism, and also diet) as they relate to farmed animals. And we need to find scalable interventions that can lead to meaningful shifts in the areas mentioned above.

Personally, as above, I see a lot of “institutional” efforts as having quite large individual components: if we want someone to sign a petition, that’s an individual ask in service of an institutional effort. So too with many fundraising asks, voting, et cetera. As I mention below, it seems that only in the case of litigating existing legislation we’re free of dealing directly with individuals in some way.

At least in the US, attitudes and intentions towards animals seem quite good [….] These numbers could certainly be higher, but overall they seem decent and have been at least somewhat validated with the success of state-level legislation passed at the poll.

I agree that these numbers look good, but I’d also say that California is probably also low-hanging fruit for behavioural change at this point. They lead the country in not only animal protection laws, but also clean air (through CARB under the CAA) and things like warning about heavy metals like lead in consumer goods (Prop 65).

In order to effect change through a popular vote, it looks like 21 states have initiated, non-constitutional statutes. I’m sure you and others know a lot more about this than I do, and I haven’t seen state numbers, but I’m somewhat doubtful that a good number of the included red states would pass something similar to Prop 12, in part because political beliefs are related to attitudes towards animals. In addition, even though Ohio is a lot closer to the political center than a number of the other statute states, it’s also one of the largest egg-producing states in the country. I’m guessing there would be a very strong pushback to defend that industry and its exports to other states and countries.

That said, the blue states up the West Coast seem to be very good candidates. And as above, if ballot initiatives have a decent chance of passing, then by all means, we should be doing that work! If we don’t have that support, there is other work that needs to be done.

Of course, the attitude-behavior and citizen-consumer gaps mean attitudes and intentions don’t necessarily translate to other behavior changes, like reduced animal product consumption or choosing cage-free over caged eggs. But to me, this suggests we need to focus on measuring and changing the behaviors we’re interested in, rather than the attitudes and intentions that don’t translate.

As I mention in the piece, there is indeed a well-known gap between each of attitudes, intentions, and behavior.

That does not mean that there is zero correlation between the three--and we don’t get near-perfect translations when dealing with people like we do in other sciences. So in some fields, you do X, and 97.3% of the variance in Y is explained. Dealing with people is much messier than that, unfortunately. You necessarily get larger error terms in studies where perfect control isn’t possible (i.e., we can’t randomly assign people to be raised a certain way).

This doesn’t mean that there isn’t still a meaningful association. It just means that the relationships are smaller because of other noise. This may be frustrating, but we really have no other choice: As hard as it is to change behavior in general, it is generally even harder to do when we don’t have those supportive attitudes and intentions in place already.

More broadly, I agree with a commenter elsewhere that this discussion seems to focus primarily on corporate animal welfare campaigns when there are many other institutional avenues to consider.

That’s a fair point, and one we should have made clearer in the post. That said, aside from litigation where the laws are already in place, I would stand by my points about considering the entire ecosystem of individual and institution.

(Lastly, at the risk of nit picking on a very minor note in the intro, but social sciences like economics and political science often consider institutions, so I’m not sure social science necessarily inclines one toward the individual.)

Fair enough, and thanks for the correction--I shouldn't have over-generalized. That said, it was a section where we were owning up to our bias, so I hope that additional bias in how I think about the social sciences can be included in that as well.

Comment by tom_beggs on Reality is often underpowered · 2019-11-27T18:05:31.294Z · score: 5 (2 votes) · EA · GW

Thanks for the post, Greg!

I think there are a couple of things worth mentioning that may allay some of the "small sample" cases somewhat. It's true, for example, that we generally never have the ability to randomize countries into treatments, and that there simply aren't enough countries to really test study designs that have a large number of conditions. However, we have some other options in our pocket for cases such as those. If we don’t really need to stay at the country level, we can do things like zoom in and randomize quite large groups of people within the country of interest. We can also use techniques like multilevel modeling to look at the effects at both levels at the same time. We also have quasi-experimental designs: in smoking cessation campaigns, for instance, researchers have used a pre-post design for each country or region: see what the baseline tobacco use is in that region, introduce your campaign, and see how smoking changes.

Now, this is not a perfect design. The country hasn't been randomly assigned and there's no control group. And we have the threats of maturation as well: what if people in that country would have cut back for another reason even without our campaign? That said, when we see a positive result in one country, and then two, and then five, ten, and fifteen different countries, we gain more confidence that it is not simply spurious. We can also compare what's happening in the campaign countries to somewhat similar countries that did not get the intervention across the same timeframe. This is not a perfect control group. But it can help us to have more confidence in what we're seeing.

So with this one counter-example, I’m basically arguing that we shouldn’t be thinking "RCT or bust." Reality is simply too messy. But we still have a large number of tools at our disposal that will give us very good information. It's not perfect information, but it's the best we can do. And we can learn a whole lot from it.

(Another point worth making is that we can use meta-analyses to help determine what the true effect size may be across a number of smaller studies. Each study on its own may be under-powered, but if we have even five or ten of them, we can get a much, much better estimate. This approach can also help control for things like regional differences and failures of randomization.)

So as Will mentions, I think we should be working on a case-by-case basis to determine what the strongest possible research design would be in each case. We should also weigh the cost of collecting that best-case evidence against the cost of other possible research designs in order to find the right blend of methodological rigor and real-world practicality.