<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
	
	>
<channel>
	<title>
	Comments for Spencer Greenberg	</title>
	<atom:link href="https://www.spencergreenberg.com/comments/feed/" rel="self" type="application/rss+xml" />
	<link>https://www.spencergreenberg.com</link>
	<description></description>
	<lastBuildDate>Wed, 21 Oct 2026 22:26:31 +0000</lastBuildDate>
	<sy:updatePeriod>
	hourly	</sy:updatePeriod>
	<sy:updateFrequency>
	1	</sy:updateFrequency>
	<generator>https://wordpress.org/?v=7.1.4</generator>
	<item>
		<title>
		Comment on Thought Experiments: Very Valuable, But Also Tricky by Spencer		</title>
		<link>https://www.spencergreenberg.com/2020/11/thought-experiments-very-valuable-but-also-tricky/#comment-71499</link>

		<dc:creator><![CDATA[Spencer]]></dc:creator>
		<pubDate>Wed, 21 Oct 2026 22:26:17 +0000</pubDate>
		<guid isPermaLink="false">https://www.spencergreenberg.com/?p=5008#comment-71499</guid>

					<description><![CDATA[In reply to &lt;a href=&quot;https://www.spencergreenberg.com/2020/11/thought-experiments-very-valuable-but-also-tricky/#comment-71489&quot;&gt;Jerry&lt;/a&gt;.

Thanks for the feedback! I updated the example to make it better.]]></description>
			<content:encoded><![CDATA[<p>In reply to <a href="https://www.spencergreenberg.com/2020/11/thought-experiments-very-valuable-but-also-tricky/#comment-71489">Jerry</a>.</p>
<p>Thanks for the feedback! I updated the example to make it better.</p>
]]></content:encoded>
		
			</item>
		<item>
		<title>
		Comment on Thought Experiments: Very Valuable, But Also Tricky by Jerry		</title>
		<link>https://www.spencergreenberg.com/2020/11/thought-experiments-very-valuable-but-also-tricky/#comment-71489</link>

		<dc:creator><![CDATA[Jerry]]></dc:creator>
		<pubDate>Wed, 21 Oct 2026 16:34:52 +0000</pubDate>
		<guid isPermaLink="false">https://www.spencergreenberg.com/?p=5008#comment-71489</guid>

					<description><![CDATA[I&#039;m here from the clearer thinking newsletter, but I think your example for #2 is not incoherent. If anything it is a good example for #1. &quot;As much like yourself as possible, but a chicken&quot; may simply be not at all like yourself! That doesn&#039;t mean it&#039;s incoherent, simply pointless to add to the thought experiment and extremely difficult to imagine. A better one would be &quot;To the best of your knowledge, do you think that a chicken in a factory farm would rather not exist, if we were able to give them the choice and they were able to understand the choice?&quot;, or even, &quot;given what you know about chicken cognition and factory farming, do you think a given chicken is better off not existing than being in a factory farm?&quot; 

That said, for me it&#039;s an easy answer, based on everything I know about chickens and their cognition, and everything I know about standard practices in factory farms, I would much rather not exist than be a chicken in one. Empathy is still possible for minds that are very different to ours.]]></description>
			<content:encoded><![CDATA[<p>I&#8217;m here from the clearer thinking newsletter, but I think your example for #2 is not incoherent. If anything it is a good example for #1. &#8220;As much like yourself as possible, but a chicken&#8221; may simply be not at all like yourself! That doesn&#8217;t mean it&#8217;s incoherent, simply pointless to add to the thought experiment and extremely difficult to imagine. A better one would be &#8220;To the best of your knowledge, do you think that a chicken in a factory farm would rather not exist, if we were able to give them the choice and they were able to understand the choice?&#8221;, or even, &#8220;given what you know about chicken cognition and factory farming, do you think a given chicken is better off not existing than being in a factory farm?&#8221; </p>
<p>That said, for me it&#8217;s an easy answer, based on everything I know about chickens and their cognition, and everything I know about standard practices in factory farms, I would much rather not exist than be a chicken in one. Empathy is still possible for minds that are very different to ours.</p>
]]></content:encoded>
		
			</item>
		<item>
		<title>
		Comment on Why We Need And Lose Institutions by David Harold Chester		</title>
		<link>https://www.spencergreenberg.com/2026/07/why-we-need-and-lose-institutions/#comment-70838</link>

		<dc:creator><![CDATA[David Harold Chester]]></dc:creator>
		<pubDate>Wed, 23 Sep 2026 08:38:43 +0000</pubDate>
		<guid isPermaLink="false">https://www.spencergreenberg.com/?p=4998#comment-70838</guid>

					<description><![CDATA[Great! 
The institution that we need to replace, having lost it too, is about how our social business system of macroeconomics works (after discovering of what it consists). This has as history where it was badly treated as a pseudo-science that was based on intuition and biased claims. Instead it should be based on logic after it has its terms properly defined and its necessary assumptions clearly stated and confirmed.

We need this new science in order that governments are provided with better ways for investigating the effects of making policy changes by simulation of the whole social business system to find the subsequent results. 

This has already been tried, but in every case there have been limitations due to not being able to properly represent the whole of any national economy. So the results are biased to support specific political needs. They are not sufficiently general to cover all of the situation and balance the general needs of the whole population. In connection with the deliberate establishment of non-availability of useful sites of land, it can be shown that this results in social inequalities and greater poverty.]]></description>
			<content:encoded><![CDATA[<p>Great!<br />
The institution that we need to replace, having lost it too, is about how our social business system of macroeconomics works (after discovering of what it consists). This has as history where it was badly treated as a pseudo-science that was based on intuition and biased claims. Instead it should be based on logic after it has its terms properly defined and its necessary assumptions clearly stated and confirmed.</p>
<p>We need this new science in order that governments are provided with better ways for investigating the effects of making policy changes by simulation of the whole social business system to find the subsequent results. </p>
<p>This has already been tried, but in every case there have been limitations due to not being able to properly represent the whole of any national economy. So the results are biased to support specific political needs. They are not sufficiently general to cover all of the situation and balance the general needs of the whole population. In connection with the deliberate establishment of non-availability of useful sites of land, it can be shown that this results in social inequalities and greater poverty.</p>
]]></content:encoded>
		
			</item>
		<item>
		<title>
		Comment on Don&#8217;t Use The Baseline Value As One Of The Predictors When You&#8217;re Predicting A Change In The Same Variable by Ante		</title>
		<link>https://www.spencergreenberg.com/2025/12/dont-use-the-baseline-value-as-one-of-the-predictors-when-youre-predicting-a-change-in-the-same-variable/#comment-70169</link>

		<dc:creator><![CDATA[Ante]]></dc:creator>
		<pubDate>Wed, 26 Aug 2026 13:38:47 +0000</pubDate>
		<guid isPermaLink="false">https://www.spencergreenberg.com/?p=4636#comment-70169</guid>

					<description><![CDATA[In reply to &lt;a href=&quot;https://www.spencergreenberg.com/2025/12/dont-use-the-baseline-value-as-one-of-the-predictors-when-youre-predicting-a-change-in-the-same-variable/#comment-70168&quot;&gt;Spencer&lt;/a&gt;.

Hi Spencer, thanks for the clarification.

Cheers!]]></description>
			<content:encoded><![CDATA[<p>In reply to <a href="https://www.spencergreenberg.com/2025/12/dont-use-the-baseline-value-as-one-of-the-predictors-when-youre-predicting-a-change-in-the-same-variable/#comment-70168">Spencer</a>.</p>
<p>Hi Spencer, thanks for the clarification.</p>
<p>Cheers!</p>
]]></content:encoded>
		
			</item>
		<item>
		<title>
		Comment on Don&#8217;t Use The Baseline Value As One Of The Predictors When You&#8217;re Predicting A Change In The Same Variable by Spencer		</title>
		<link>https://www.spencergreenberg.com/2025/12/dont-use-the-baseline-value-as-one-of-the-predictors-when-youre-predicting-a-change-in-the-same-variable/#comment-70168</link>

		<dc:creator><![CDATA[Spencer]]></dc:creator>
		<pubDate>Wed, 26 Aug 2026 13:35:18 +0000</pubDate>
		<guid isPermaLink="false">https://www.spencergreenberg.com/?p=4636#comment-70168</guid>

					<description><![CDATA[In reply to &lt;a href=&quot;https://www.spencergreenberg.com/2025/12/dont-use-the-baseline-value-as-one-of-the-predictors-when-youre-predicting-a-change-in-the-same-variable/#comment-70156&quot;&gt;Ante Sch&lt;/a&gt;.

Hi Ante. Sorry for the confusion - I was not saying that this isn&#039;t a problem if (2) isn&#039;t met; I was saying that it&#039;s definitely a problem if (1) and (2) are met (if (1) and (2) -&gt; problem, not: lack of (2) -&gt; not problem). I updated the language for clarity. 

This issue can be avoided by predicting the final outcome (not the change in that outcome) using the base value as an independent variable. Of course, this won&#039;t settle WHY the outcome changed (e.g., you can&#039;t tell if it&#039;s regression to the mean, placebo, effect of the intevention itself, or something else), but it can tell that it did change (unlike the approach I critique in this article, where you predict the change in outcome and use the base value as an independent variable).]]></description>
			<content:encoded><![CDATA[<p>In reply to <a href="https://www.spencergreenberg.com/2025/12/dont-use-the-baseline-value-as-one-of-the-predictors-when-youre-predicting-a-change-in-the-same-variable/#comment-70156">Ante Sch</a>.</p>
<p>Hi Ante. Sorry for the confusion &#8211; I was not saying that this isn&#8217;t a problem if (2) isn&#8217;t met; I was saying that it&#8217;s definitely a problem if (1) and (2) are met (if (1) and (2) -> problem, not: lack of (2) -> not problem). I updated the language for clarity. </p>
<p>This issue can be avoided by predicting the final outcome (not the change in that outcome) using the base value as an independent variable. Of course, this won&#8217;t settle WHY the outcome changed (e.g., you can&#8217;t tell if it&#8217;s regression to the mean, placebo, effect of the intevention itself, or something else), but it can tell that it did change (unlike the approach I critique in this article, where you predict the change in outcome and use the base value as an independent variable).</p>
]]></content:encoded>
		
			</item>
		<item>
		<title>
		Comment on Don&#8217;t Use The Baseline Value As One Of The Predictors When You&#8217;re Predicting A Change In The Same Variable by Ante Sch		</title>
		<link>https://www.spencergreenberg.com/2025/12/dont-use-the-baseline-value-as-one-of-the-predictors-when-youre-predicting-a-change-in-the-same-variable/#comment-70156</link>

		<dc:creator><![CDATA[Ante Sch]]></dc:creator>
		<pubDate>Tue, 25 Aug 2026 20:32:03 +0000</pubDate>
		<guid isPermaLink="false">https://www.spencergreenberg.com/?p=4636#comment-70156</guid>

					<description><![CDATA[Hi Spencer,

I came across a video of yours and have spent the last few hours going through your site with a lot of interest. I work on psychological interventions, and your post on using the baseline as a predictor of change hit a nerve, so let me think out loud for a moment.

The first thing it sharpened for me is the self selection angle. People tend to enrol in intervention studies at their worst, so even with no treatment at all you would expect their scores to come down at follow up, purely through regression to the mean. That alone should make us cautious about a lot of pre and post designs.

Where I wanted to push back was on your condition (2), that there is no mechanism by which a higher baseline drives a larger change. In symptom research it felt to me like something like that is almost always in play. Patients with higher initial scores have more room to move on a bounded scale, and there are real clinical reasons to think severity can moderate how much someone responds. So my first instinct was that condition (2) is routinely violated, and that this is a separate bias one would have to account for.

But following it through, I am no longer sure that helps the way I first thought. The measurement noise effect, the bounded scale room to move effect, and any genuine severity moderation all push in the same direction, higher baseline and bigger drop. So if I regress change on baseline, the coefficient is a blend of all three and I cannot tell which is which. If anything that seems to reinforce your point rather than complicate it.

So here is my actual question. Given that these sources all share a sign and stack on top of each other, is there any design or model that genuinely separates a real severity by treatment effect from the artefacts, short of a randomised control group with a baseline by treatment interaction term? Or is the honest answer that in observational intervention data they are simply not separable?

Thanks again for the post!

Best,
Ante]]></description>
			<content:encoded><![CDATA[<p>Hi Spencer,</p>
<p>I came across a video of yours and have spent the last few hours going through your site with a lot of interest. I work on psychological interventions, and your post on using the baseline as a predictor of change hit a nerve, so let me think out loud for a moment.</p>
<p>The first thing it sharpened for me is the self selection angle. People tend to enrol in intervention studies at their worst, so even with no treatment at all you would expect their scores to come down at follow up, purely through regression to the mean. That alone should make us cautious about a lot of pre and post designs.</p>
<p>Where I wanted to push back was on your condition (2), that there is no mechanism by which a higher baseline drives a larger change. In symptom research it felt to me like something like that is almost always in play. Patients with higher initial scores have more room to move on a bounded scale, and there are real clinical reasons to think severity can moderate how much someone responds. So my first instinct was that condition (2) is routinely violated, and that this is a separate bias one would have to account for.</p>
<p>But following it through, I am no longer sure that helps the way I first thought. The measurement noise effect, the bounded scale room to move effect, and any genuine severity moderation all push in the same direction, higher baseline and bigger drop. So if I regress change on baseline, the coefficient is a blend of all three and I cannot tell which is which. If anything that seems to reinforce your point rather than complicate it.</p>
<p>So here is my actual question. Given that these sources all share a sign and stack on top of each other, is there any design or model that genuinely separates a real severity by treatment effect from the artefacts, short of a randomised control group with a baseline by treatment interaction term? Or is the honest answer that in observational intervention data they are simply not separable?</p>
<p>Thanks again for the post!</p>
<p>Best,<br />
Ante</p>
]]></content:encoded>
		
			</item>
		<item>
		<title>
		Comment on 10 Structures of Problems by Martin Michaels		</title>
		<link>https://www.spencergreenberg.com/2020/08/10-structures-of-problems/#comment-69883</link>

		<dc:creator><![CDATA[Martin Michaels]]></dc:creator>
		<pubDate>Fri, 14 Aug 2026 09:45:13 +0000</pubDate>
		<guid isPermaLink="false">https://www.spencergreenberg.com/?p=4890#comment-69883</guid>

					<description><![CDATA[Another problem structure: &quot;death by a thousand paper cuts&quot;
The accumulation of small annoyances and inefficiencies that, individually are too small to fix or notice, but which together compound in a way that exhausts resources, degrades performance, or destroys morale.]]></description>
			<content:encoded><![CDATA[<p>Another problem structure: &#8220;death by a thousand paper cuts&#8221;<br />
The accumulation of small annoyances and inefficiencies that, individually are too small to fix or notice, but which together compound in a way that exhausts resources, degrades performance, or destroys morale.</p>
]]></content:encoded>
		
			</item>
		<item>
		<title>
		Comment on Ten Skills Our Civilization Needs by Greg Flint		</title>
		<link>https://www.spencergreenberg.com/2020/10/ten-skills-our-civilization-needs/#comment-69828</link>

		<dc:creator><![CDATA[Greg Flint]]></dc:creator>
		<pubDate>Wed, 12 Aug 2026 04:01:10 +0000</pubDate>
		<guid isPermaLink="false">https://www.spencergreenberg.com/?p=4948#comment-69828</guid>

					<description><![CDATA[I think number 11 (or possibly even 1) would be face-to-face collaboration with &quot;others&quot;. Our electoral (competitive, binary) oligarchy that we have as a governance system, and which we erroneously call &quot;democracy&quot;, pushes us further and further into tribal groups and echo chambers.

There is one way out of this, which is to give genuine democracy a try (arroding to the original meaning of the word, not its misuse over the last couplf of hundred years). Wherever it&#039;s tried around the world, which has been happening increasingly over the last 20 years, it is incredibly successful. It even enabled the most Catholic country in the world - the Republic of Ireland - to find a way to reduce the numbers of deaths and injuries resulting from their outlawing of abortion under any circumstances in 1983.]]></description>
			<content:encoded><![CDATA[<p>I think number 11 (or possibly even 1) would be face-to-face collaboration with &#8220;others&#8221;. Our electoral (competitive, binary) oligarchy that we have as a governance system, and which we erroneously call &#8220;democracy&#8221;, pushes us further and further into tribal groups and echo chambers.</p>
<p>There is one way out of this, which is to give genuine democracy a try (arroding to the original meaning of the word, not its misuse over the last couplf of hundred years). Wherever it&#8217;s tried around the world, which has been happening increasingly over the last 20 years, it is incredibly successful. It even enabled the most Catholic country in the world &#8211; the Republic of Ireland &#8211; to find a way to reduce the numbers of deaths and injuries resulting from their outlawing of abortion under any circumstances in 1983.</p>
]]></content:encoded>
		
			</item>
		<item>
		<title>
		Comment on Three Types of Nuanced Thinking by Michelle Donelson		</title>
		<link>https://www.spencergreenberg.com/2020/06/three-types-of-nuanced-thinking/#comment-69821</link>

		<dc:creator><![CDATA[Michelle Donelson]]></dc:creator>
		<pubDate>Tue, 11 Aug 2026 22:15:25 +0000</pubDate>
		<guid isPermaLink="false">https://www.spencergreenberg.com/?p=1685#comment-69821</guid>

					<description><![CDATA[Everyone should read this, if someone just took the time to see the difference between me and they, or us and them.  I do not have a predictable outcome, I am not who you think I am.]]></description>
			<content:encoded><![CDATA[<p>Everyone should read this, if someone just took the time to see the difference between me and they, or us and them.  I do not have a predictable outcome, I am not who you think I am.</p>
]]></content:encoded>
		
			</item>
		<item>
		<title>
		Comment on 10 Structures of Problems by Martin Michaels		</title>
		<link>https://www.spencergreenberg.com/2020/08/10-structures-of-problems/#comment-69738</link>

		<dc:creator><![CDATA[Martin Michaels]]></dc:creator>
		<pubDate>Fri, 07 Aug 2026 12:54:19 +0000</pubDate>
		<guid isPermaLink="false">https://www.spencergreenberg.com/?p=4890#comment-69738</guid>

					<description><![CDATA[Are you collecting new ones?

Because I have one which maybe could be called a &quot;tangled knot&quot; although I&#039;m not sure that&#039;s the best description.  It&#039;s when two or more other problems produce a new emergent issue, like the whole is more than the sum of its parts.]]></description>
			<content:encoded><![CDATA[<p>Are you collecting new ones?</p>
<p>Because I have one which maybe could be called a &#8220;tangled knot&#8221; although I&#8217;m not sure that&#8217;s the best description.  It&#8217;s when two or more other problems produce a new emergent issue, like the whole is more than the sum of its parts.</p>
]]></content:encoded>
		
			</item>
	</channel>
</rss>
