A summary of scientific method

By Peter Kosso.

This book is crap, stay away from it. It’s very short, which was the only reason why I actually read it cover to cover. Kosso neglects some very important points you’d want to see in a publication like this; on the list of recommended reading he includes Kuhn but not Popper, and Popper’s name isn’t even mentioned. Presumably because he disagrees with Popper about the importance of falsification. Conceptually he doesn’t talk about and doesn’t seem to understand how crucial is the requirement in science that you restrict the (potential) outcome space when forming hypotheses. He picks out history and archaeology as examples of ‘social sciences’; maybe because that’s the closest he’s ever been to the social sciences? He talks about how experimental designs can play a role here, but doesn’t include a single word about the role of statistics in scientific disciplines.

I’d probably give it 1 out of 5 on amazon. He reads as if he doesn’t have a clue. The only good thing about the book is that it is quite short.

 

Richard Feynman: Fun to Imagine | Using physics to explain how the world works

This is great stuff:

Stuff

i. I was considering covering this study in a bit more detail, but I decided against it because workplace filters probably would not like it very much – it would contain words such filters do not like (no, I’m not thinking of words like ‘sociodemographic characteristics’ or ‘multiple regression analyses’). I know a few people sometimes read my blog from work and if you’re one of them, let me just say that you should probably not read this while at work.

…

ii. Population Trends in the Incidence and Outcomes of Acute Myocardial Infarction

“The age- and sex-adjusted incidence of myocardial infarction increased from 274 cases per 100,000 person-years in 1999 to 287 cases per 100,000 person-years in 2000, and it decreased each year thereafter, to 208 cases per 100,000 person-years in 2008, representing a 24% relative decrease over the study period. [...]

The proportion of patients who underwent revascularization within 30 days after myocardial infarction increased from 40.7% in 1999 to 47.2% in 2008 (P<0.001 for trend). Among patients with ST-segment elevation myocardial infarction, 49.9% underwent revascularization in 1999 as compared with 69.6% in 2008 (P<0.001 for trend). Among patients with non–ST-segment elevation myocardial infarction, 33.4% underwent revascularization in 1999 as compared with 41.3% in 2008 (P<0.001 for trend) [...]

The proportion of patients with myocardial infarction who were known to have undergone troponin I testing increased from 53% in 1999 to 84% in 2004, with stable testing rates between 2004 and 2008. [...]

The age- and sex-adjusted 30-day mortality after myocardial infarction decreased from 10.5% in 1999 to 7.8% in 2008 (P<0.001 for linear trend). This decrease was driven by the case fatality rate for non–ST-segment elevation myocardial infarction, which decreased from 10.0% to 7.6% (P<0.001 for trend); there was no significant change over time for ST-segment elevation myocardial infarction (P = 0.81). The multivariable adjusted odds ratio for death at 30 days after myocardial infarction was 0.76 (95% confidence interval [CI], 0.65 to 0.89) in 2008 as compared with 1999.”

Short version: Fewer people got a(n ST-segment elevation) myocardial infarction even though more people were subjected to fancy testing, more people got access to fancy treatment, and the people in the sample who got a non-ST-segment MI during the study period were less likely to die from it. But…

“observed reductions in case fatality rates could be attributable to secular trends in ascertainment of myocardial infarction and decreased severity on presentation, as well as any improvements in management of acute myocardial infarction.44 The observation that mortality after ST-segment elevation myocardial infarction (which is less influenced by the use of highly sensitive biomarkers) did not decrease over time provides support for this hypothesis.”

This could still be considered good news because if decreased severity on presentation reduces mortality it’s probably a good idea to at least have a closer look at that variable; on the other hand it’s bad news because fancy testing is expensive. Another thing:

“given the integrated medical care delivery structure in the health system that we studied and the magnitude of recent improvements in the control of risk factors within our population, our results may not be fully generalizable to other health care settings.”

Good luck finding MSM-coverage of the study including this part. I’d probably have removed the word ‘fully’. The population risk factor development during the period is a major confound.

…

iii. International migration: A panel data analysis of the determinants of bilateral flows by Anna Maria Mayda.

Click to view full size. From the paper:

“According to the international migration model, pull and push factors have either similarsized effects (with opposite signs), when migration quotas are not binding, or they both have no (or a small) effect on emigration rates, when migration quotas are binding. It is not clear, ex ante, which one of the two scenarios characterizes actual flows. Migration policies in the majority of destination countries are very restrictive, which should imply binding constraints on the number of migrants. On the other hand, even countries with binding official immigration quotas often accept unwanted (legal) immigration.8 Restrictive immigration policies are often characterized by loopholes, that leave room for potential migrants to take advantage of economic incentives. [...]

My empirical analysis also finds that inequality in the source and host economies is related to the size of emigration rates as predicted by Borjas (1987) selection model. An increase in the origin country’s relative inequality has a non-monotonic effect on the size of the emigration rate: the impact is estimated to be positive if there is positive selection, negative if there is negative selection. Among the variables affecting the costs of migration, distance between destination and origin countries appears to be the most important one: Its effect is negative, significant and steady across specifications. On the other hand, there is no evidence that cultural variables related to each country pair play a significant role. Demographics – in particular, the share of the origin country’s population who is young – shape bilateral flows as predicted by the theory. Since the effect of geography and demographics works through the supply side of the model, their impact should be even stronger when migration quotas are relaxed, which is what I find in the data. [...]

Since immigrants are likely to receive support from other immigrants from the same origin country already established in the host country, they will have an incentive to choose destinations with larger communities of fellow citizens. Network effects imply that bilateral migration flows are highly correlated over time, which is what the data shows.”

…

iv. Via npr:

“It’s a sound you would never want to hear in real life, but this a safe way to eavesdrop. Just one warning: For the first two minutes of this video, nothing happens, nothing I could hear, anyway. Then there’s a countdown, and at 2:24 from the top … the bomb bursts; at 2:54 the blast hits.”

…

v. Does Thinking Really Hard Burn More Calories? Interesting piece. Unfortunately(?), “for most people, the body easily supplies what little extra glucose the brain needs for additional mental effort.”

I would be very interested in seeing a study on this including type 1 diabetics. Hard thinking for extended periods of time – like, say, a four-hour chess game or an exam – impacts my blood glucose in a very significant way; it drops like a stone if I don’t take precautions. This is despite the fact that hard thinking under such circumstances is often, as mentioned in the article, linked to stress and the release of cortisol, one of the primary functions of which is to increase blood sugar.

…

vi. TV from a different world:

A few more physics lectures

Given that I’ve already posted two posts about Muller’s lectures, I’m not posting this because I think you’d otherwise not be able to find these – it’s more of a relatively effortless ‘stuff I’m (also) doing’/'worth remembering’/'I guess I should keep updating the blog even though almost nobody is reading along at this time of year’-post.

Scientific knowledge across countries

I’m sure I’ve seen some of this stuff before (Razib Khan may have covered it), but I’m pretty sure I have not blogged it. Link to the source here, click to view the figures/tables in full size:

Of course formal education matters, a lot:

Interestingly, the link also has data related to a recent post:

It seems that public opinion doesn’t change very much over time. I thought this last one was interesting (if anyone knows of any related Danish data, let me know in the comment section):

Note that this is only the “very great prestige”-proportion, so there may be stuff going on we don’t know about. Note how much both ‘teacher’ and ‘military officer’ has changed over time. Something funny may be going on here; ‘farmer’ is more prestigious than ‘Member of Congress’ and ‘Lawyer’ (‘well of course it is,’ you might say, but…).

A few more physics lectures

I’ve spent most of the day watching videos like these (I’ve now seen the first 12 lectures). He rambles a bit sometimes, but it’s youtube – it’s easy to skip stuff if you don’t want to watch him spend several minutes talking about his hybrid car and how awesome it is (or whatever). Unless you’re a physics major – in which case you’re not in the target group anyway – I’d be willing to bet there’s stuff covered in these lectures you didn’t know. Stuff you didn’t know you didn’t know, because you didn’t think about it.

Some physics lectures and a bit of other stuff

i. A few physics lectures from Berkeley’s youtube channel. As they put it in the description: “The most interesting and important topics in physics, stressing conceptual understanding rather than math, with applications to current events.” I think it’s awesome that stuff like this is available for free for you to watch whenever you want to:

Note that there are a lot of lectures available here (26, each lasting about ~70 minutes).

ii. I may have blogged this before, but I don’t think so. Anyway this is just so cute you’ll not be harmed by watching it again in case you already have:

iii. “There’s also a long history of seemingly rational scientists who were willing to sacrifice their physical comfort, as well as their lives, for the sake of knowledge.” Keyword: Seemingly.

I won’t comment on the others, but mr. Hill’s experiment as described in that article could in my opinion aptly be categorized as insane (‘do the same thing over and over and expect different results…’).

iv. “before the development of modern prison systems, the death penalty was also used as a generalised form of punishment. During the reign of Henry VIII, as many as 72,000 people are estimated to have been executed.^[26] By 1820 in Britain, there were 160 crimes that were punishable by death, including crimes such as shoplifting, petty theft, stealing cattle, or cutting down trees in public place.” (more)

Science does more harm than good?

I just had to share this (US data, from the GSS):

Razib doesn’t comment, he just gives you the data. I won’t comment much on the above graph now either, but that’s only because I’m dumbstruck right now – I have just no idea what to even say to all those people who do not strongly disagree.

‘Publish or perish’ and bias

Do Pressures to Publish Increase Scientists’ Bias? An Empirical Support from US States Data, by Daniele Fanelli (link). Abstract:

“The growing competition and “publish or perish” culture in academia might conflict with the objectivity and integrity of research, because it forces scientists to produce “publishable” results at all costs. Papers are less likely to be published and to be cited if they report “negative” results (results that fail to support the tested hypothesis). Therefore, if publication pressures increase scientific bias, the frequency of “positive” results in the literature should be higher in the more competitive and “productive” academic environments. This study verified this hypothesis by measuring the frequency of positive results in a large random sample of papers with a corresponding author based in the US. Across all disciplines, papers were more likely to support a tested hypothesis if their corresponding authors were working in states that, according to NSF data, produced more academic papers per capita. The size of this effect increased when controlling for state’s per capita R&D expenditure and for study characteristics that previous research showed to correlate with the frequency of positive results, including discipline and methodology. Although the confounding effect of institutions’ prestige could not be excluded (researchers in the more productive universities could be the most clever and successful in their experiments), these results support the hypothesis that competitive academic environments increase not only scientists’ productivity but also their bias. The same phenomenon might be observed in other countries where academic competition and pressures to publish are high.”

Figure 2:

An important bit on ‘”negative” results’ from the paper:

“Words like “positive”, “significant”, “negative” or “null” are common scientific jargon, but are obviously misleading, because all results are equally relevant to science, as long as they have been produced by sound logic and methods [11,12]. Yet, literature surveys and meta-analyses have extensively documented an excess of positive and/or statistically significant results in fields and subfields of, for example, biomedicine [13], biology [14], ecology and evolution [15], psychology [16], economics [17], sociology [18].
Many factors contribute to this publication bias against negative results, which is rooted in the psychology and sociology of science. Like all human beings, scientists are confirmationbiased (i.e. tend to select information that supports their hypotheses about the world) [19,20,21], and they are far from indifferent to the outcome of their own research: positive results make them happy and negative ones disappointed [22]. This bias is likely to be reinforced by a positive feedback from the scientific community. Since papers reporting positive results attract more interest and are cited more often, journal editors and peer reviewers might tend to favour them, which will further increase the desirability of a positive outcome to researchers, particularly if their careers are evaluated by counting the number of papers listed in their CVs and the impact factor of the journals they are published in.
Confronted with a “negative” result, therefore, a scientist might be tempted to either not spend time publishing it (what is often called the “file-drawer effect”, because negative papers are imagined to lie in scientists’ drawers) or to turn it somehow into a positive result. This can be done by re-formulating the hypothesis (sometimes referred to as HARKing: Hypothesizing After the Results are Known [23]), by selecting the results to be published [24], by tweaking data or analyses to “improve” the outcome, or by willingly and consciously falsifying them [25]. Data fabrication and falsification are probably rare, but other questionable research practices might be relatively common [26].”

Earth (some stuff you should know)

I had an interesting discussion yesterday which touched briefly upon a few of these subjects, so I decided to take a closer look at the data just to make sure I wasn’t completely wrong about the stuff I thought I knew – and now I’m glad I did as I seem to have somehow picked up a mistaken idea about the land area of the Southern Hemisphere (I thought it was even smaller than it is). Now, if you asked a random guy he wouldn’t know most of these numbers or even the relevant neighbourhood. Somehow I feel like people should. So here we go, most of these numbers are pulled from wikipedia:

1. Asia covers 8.7 % of the Earth’s total surface area and hosts ~60 % of the world’s current human population. It covers 29.5 % of the land area of Earth.

1a. Africa covers 6 % of the Earth’s total surface area and hosts ~14-15 % of the world’s population. It covers 20.4% of the total land area.

1b. North America: 4.8 % of surface area, 8 % of population. 16.5 % of total land area.

1c. South America: 3.5 % of surface area, 6 % of population. 12.0 % of total land area.

1d. Antarctica: 2.7 % of surface area, 0 % of population. 9.2% of total land area.

1e. Europe: 2 % of surface area, 11.5 % of population. 6.8 % of total land area.

1f. Australia: 1.5 % of surface area, 0.5 % of population. 5.1 % of total land area.

2. Russia covers 17,075,400 square kilometres. Europe and Australia combined make out ~17,8 mio. square kilometres, a number which incidentally is about the same as South America. So if we for a moment disregard the fact that Russia already makes up 40 % of the total area of Europe, it’s large enough to almost cover the two smallest continents combined.

3. According to a 2010 census, the population of China was/is 1,339,724,852 – which is more than 19 % of the population of Earth. This is a higher population than that of any single continent which is not Asia. The population of China is significantly larger than the combined populations of South America (385,7 mio), North America (529 mio) and Australia (31,26). It’s larger than the combined populations of Europe and North America. Here’s a neat image comparing sizes and populations of the continents.

4. This source notes that: “In the Northern Hemisphere, the ratio of land to ocean is about 1 to 1.5. The ratio of land to ocean in the Southern Hemisphere is 1 to 4.” Translating those ratios into percentages of the hemispheres, it turns out that in the Northern Hemisphere 60 % of the area is made up of ocean and 40 % is covered by land, whereas only 20 % of the Southern Hemisphere is covered by land and 80 % is covered by ocean. Oceans cover roughly 70,8 % of the total area of earth and land masses cover 29,2 %, so these numbers are probably ok. Here’s an image from Wikipedia:

About 90 percent of the human population lives on the Northern Hemisphere – the combined human population of the entire Southern Hemisphere is smaller than the population of Europe.

4a. The Pacific Ocean covers a larger area than all land masses of Earth combined.

4b. The Atlantic Ocean covers as a very rough approximation the same area (106 mio. square kilometres) as the total land area of the Northern Hemisphere. It covers an area corresponding to more than 70 percent of the total land area of earth.

4c. The Indian Ocean covers 68,556,000 square kilometres, approximately the same area as Asia and North America combined.

4d. The average depth of the world oceans is about 3.8 kilometers (link).

5. I can’t copy the image, but go here for a really neat illustration of the surface elevation of the areas of Earth – I’m really annoyed I can’t copy this and put it in the post. Antarctica has by far the highest mean elevation of all continents. According to this source, the mean elevation of the continent is 2,286 m. Disregarding Antarctica (which can be considered somewhat an outlier because of the ice-thing), it seems that there’s a connection between the area of a continent and its mean elevation – i.e. the larger the area of the continent, the higher the elevation. Here’s a relevant paper.)

The neutrino velocity thing

Here’s the paper, go have fun if you’re into that kind of thing. Here’s one non-sensationalist take on it. I love how they’ve just put all this stuff ‘out there’ for everyone to see and criticise – open science is the only kind of science worth anything.

I have naturally no idea what’s the cause of the results, but my first guess would be the existence of some measurement error they’ve been unaware of. I think it’s a bit funny how a lot of ‘average Joes’ have decided to leave comments on the badastronomy post with more or less fleshed-out ideas as to what’s driving the results, probably posted after at most five minutes of thought on the matter – given that a lot of very, very smart people have spent a lot of time doing these tests and thinking long and hard for days, weeks, months about the test designs and how to improve them. Yeah, they most likely missed some source of uncertainty in the analysis, but you’re probably not the guy who’s going to figure out what it is. Most of the Average Joes commenting aren’t Swiss patent office clerks, and he thought about the stuff for years before he ‘went public’ anyway.

Oh yeah, remember to apply the Sagan Standard here. I know the journalists covering this sure aren’t.

Khan Academy stuff of interest

Simplicity

Compare:

i.

a)

Image credit: Wikipedia.

b)

“By definition, juggling requires at least one more ball than there are hands. So, with 2 hands, omega has a minimum value of 0.5, when b = 3 and r = 1.

Newtonian physics provides some equations that apply to thrown objects:
Vh = F / f
Vv = sqrt(2 * g * H)
H = (Vv ^ 2) / 2 / g

If the throws and catches are assumed to occur at the same height, we can solve for the flight time, f, and substitute it into the other equations. As a result, we get some useful equations:
f = sqrt(8 * H / g)
Vv = g * f / 2
H = g * (f^2) / 2 = g * [(tau * omega)^2] / 8

These equations tell us that to double the flight time, you must double the initial throw velocity which means quadrupling the throw height.

Graph 1: Throw Height as a Function of tau and omega (assuming throw and catch positions at the same height, g = 9.81 m/s/s)

[...]

The Optimal Cascade
The shape and relative positions of the arcs are defined by the equations for P and the optimal arc equations. We now must determine the optimal endpoints of the arcs, which will be the throw and catch positions. We will not assume throws and catches are at the same height. Define the following new variables:
E = dwell distance = distance between catch position and throw position
theta = dwell angle

Assume we are given two parabolas in a cascade, and the values for Vh, F, and P are also given. We would like to find the exact throw and catch positions on these parabolas that minimizes the distance between throw and catch. (The equations start to get messy here, so I’ll just summarize the method and the results.) First, find the optimal value of theta to minimize the dwell distance, E. This is done by finding an equation for E in terms of the given quantities and theta, and setting the derivative with respect to theta, dE/d(theta) = 0. The result is:
tan(theta) = 2 * Vh * Vh / g / F .

Substituting this value back into equations for E and H, we get the following results:
E = P * cos(theta) .
This means the distance the hands must move, E, is slightly less than P but E is still always greater than the diameter of a ball, D. The result for H is:
H = {F + [P * sin(theta) * sin(theta)]}^2 / [4 * F * tan(theta)]
As mentioned earlier, if theta is small, H can be approximated with:
H = g * f * f / 8 .

There is a limit to how low you can make an optimal cascade. The balls from one arc must graze the balls from the other arc. The limit is reached when, at the instant each ball is thrown, it grazes the previously thrown ball from the other hand. Here is a summary of how to find the equation for this limit. First, write an equation for the distance between two consecutive throws from opposite hands. By setting the time derivative of this distance = 0, we find the time at which the distance between the balls is minimized and, in an optimal pattern, they are touching. By setting this time = 0, we get an equation for patterns where balls are grazing at the instant of the throw. The result, included for completeness, is: 2v(h)(F-E*cosϕ-D/2)+gτ^2/4(v(v)-gτ/4)=0″

(more here)

…

ii.

a)

b)

” Average linear size of areas with uniform sign of mean curvature for elastic and elasto-plastic sheets of size L/h=250, 500 and 1,000. In a crumpled state (R<0.4R0), this size describes the characteristic size of facets and ridges in the sheet. b, The total energy of elastic and elasto-plastic sheets of size L/h=250, 500 and 1,000, scaled by 1/(L/h)1/3. Transitions in the energy of the elastic sheets at R~0.75R0 and R~0.4R0 indicate the formation of a cone and the end of a single-cone regime, respectively. The plots shown are averages of three simulations, and the yield point of the elasto-plastic sheets is Y^(1/σ)=0.01." [...]

In elasto-plastic sheets, the situation is more complicated. It is evident from Fig. 3 that their average linear facet size and energy scale similarly as a function of compression to the elastic sheets. There are however differences in the two crumpling processes, which arise at early phases of crumpling. In vertices in particular, plastic deformations appear already for R/R0 close to unity. An elasto-plastic sheet is not able to transform into a cone necessary for a folding type of initial deformations, and large numbers of vertices and ridges appear soon after crumpling begins. This becomes increasingly pronounced for increasing L/h so that the relative facet diameter then decreases as shown in Fig. 3a. It is evident from this figure that the average ridge length scales now as mean(x)/L=f(L/h)g(R/R0), where function g(z) has a power-law form in a fairly large range of the argument. It is difficult to determine by simulations a functional form for f(z). The above scaling form means however that elasto-plastic sheets of different L/h can only have the same average (relative) ridge length for different degrees of compression. Consequently, the similarity of ridge patterns found for crumpled elastic sheets does not appear in elasto-plastic sheets. The lack of such similarity is also evident in the L/h-dependent distributions of the linear facet size in elasto-plastic sheets (Fig. 4d): sheet thickness must be scaled together with the other spatial dimensions to preserve the form of the distribution. The distributions of linear facet size are now well fitted by lognormal distributions, found previously for experimental facet size"

Much more here.

…

iii.

a)

b)

This and this.

…

iv.

a)

b)

(link).

(link).

“Pressure is an effect which occurs when a force is applied on a surface. Pressure is the amount of force acting on a unit area. The symbol of pressure is P [...] Mathematically:

P = F/A or P = dFn/dA

where:

P is the pressure,
F is the normal force,
A is the area.

Pressure is a scalar quantity. It relates the vector surface element (a vector normal to the surface) with the normal force acting on it. The pressure is the scalar proportionality constant that relates the two normal vectors:

dF(n) = -dA = -PndA

The minus sign comes from the fact that the force is considered towards the surface element, while the normal vector points outwards.” (link)

“Nutritional value per 100 g (3.5 oz) :
Energy 218 kJ (52 kcal)
Carbohydrates 13.81 g
- Sugars 10.39 g
- Dietary fiber 2.4 g
Fat 0.17 g
Protein 0.26 g
Water 85.56 g
Vitamin A equiv. 3 μg (0%)
Thiamine (Vit. B1) 0.017 mg (1%)
Riboflavin (Vit. B2) 0.026 mg (2%)
Niacin (Vit. B3) 0.091 mg (1%)
Pantothenic acid (B5) 0.061 mg (1%)
Vitamin B6 0.041 mg (3%)
Folate (Vit. B9) 3 μg (1%)
Vitamin C 4.6 mg (6%)
Calcium 6 mg (1%)
Iron 0.12 mg (1%)
Magnesium 5 mg (1%)
Phosphorus 11 mg (2%)
Potassium 107 mg (2%)
Zinc 0.04 mg (0%)”

(link)
…

Yeah, I know:

A big part of why we keep things simple is because otherwise we’d have died out long ago. And our brains aren’t big enough to understand all that stuff anyway, even if there were time to figure it all out, which there’s not.

But things don’t get any simpler by us trying to make them so. Even very basic stuff usually tends to be horribly complicated once you start to think about it. Once in a while you can actually convince yourself that this existence-stuff we have going on is really quite fantastic.

A glossary for research reports

Not the first of its kind, but I liked it. Link.

…

Scientific term (Actual meaning)

It has long been known that … (I haven’t bothered to look up the original reference)

…of great theoretical and practical importance (…interesting to me)

While it has not been possible to provide definite answers to these questions … (The experiments didn’t work out, but I figured I could at least get a publication out of it)

The W-Pb system was chosen as especially suitable to show the predicted behaviour. … (The fellow in the next lab had some already made up)

High-purity || Very high purity || Extremely high purity || Super-purity || Spectroscopically pure … (Composition unknown except for the exaggerated claims of the supplier)

A fiducial reference line … (A scratch)

Three of the samples were chosen for detailed study … (The results on the others didn’t make sense and were ignored)

…accidentally strained during mounting (…dropped on the floor)

…handled with extreme care throughout the experiments (…not dropped on the floor)

…Typical results are shown … (The best results are shown)

Although some detail has been lost in reproduction, it is clear from the original micrograph that … (It is impossible to tell from the micrograph)

Presumably at longer times … (I didn’t take time to find out)

The agreement with the predicted curve is excellent (fair) || good (poor) || satisfactory (doubtful) || fair (imaginary) || . . as good as could be expected (non-existent)

These results will be reported at a later date … (I might possibly get around to this sometime)

The most reliable values are those of Jones (He was a student of mine)

It is suggested that || It is believed that || It may be that … (I think)

It is generally believed that … (A couple of other guys think so too)

It might be argued that … (I have such a good answer to this objection that I shall now raise it)

It is clear that much additional work will be required before a complete understanding … (I don’t understand it)

Unfortunately, a quantitative theory to account for these effects has not been formulated … (Neither does anybody else)

Correct within an order of magnitude … (Wrong)

It is to be hoped that this work will stimulate further work in the field … (This paper isn’t very good, but neither are any of the others in this miserable subject)

Thanks are due to Joe Glotz for assistance with the experiments and to John Doe for valuable discussions … (Glotz did the work and Doe explained what it meant)

C. D. Graham, Jr., Metal. Progress 71, 75 (1957) (actual source)

…and don’t even get me started on effect sizes…

xkcd. Here’s a bit more.

The language of science

Some common phrases in scientific papers and what they really mean (here’s the link, via Ed Yong):

“It can be shown”

Somebody said they did this, but I can’t duplicate their results. I can’t even find the reference, or else I would have cited that instead. [In economics, this phrase has a lot of different meanings, depending on who the writer or lecturer is. Usually I translate that sentence simply by adding an apostrophe and a t - US]

…

“Of great theoretical and practical importance”

Means it is interesting to me or else I want it to be interesting to somebody with money so they will fund my research.

…

“The most reliable results are those obtained by Smith.”

Smith is or was my graduate research assistant.

“It is believed that…”

I think this (and either no one agrees with me or else I didn’t consult anyone).

“It is generally believed that”

I think this and at least one other person agrees with me.

…

“Additional work will be required to elucidate the mechanism”

I don’t have a clue what is going on and I’m not going to be the one to figure it out.

Stuff you should read

Dealing with the high quantity of scientific error in medicine. Many of the comments to the post are (in my opinion) uninteresting stuff about diet, but this comment is quite good, and so is Yvain’s response here. Here’s one bit from the post, Ioannidis’ corollaries:

“Corollary 1: The smaller the studies conducted in a scientific field, the less likely the research findings are to be true.

Corollary 2: The smaller the effect sizes in a scientific field, the less likely the research findings are to be true.

Corollary 3: The greater the number and the lesser the selection of tested relationships in a scientific field, the less likely the research findings are to be true.

Corollary 4: The greater the flexibility in designs, definitions, outcomes, and analytical modes in a scientific field, the less likely the research findings are to be true.

Corollary 5: The greater the financial and other interests and prejudices in a scientific field, the less likely the research findings are to be true.

Corollary 6: The hotter a scientific field (with more scientific teams involved), the less likely the research findings are to be true.”

…

It looks like this article that the first article I linked to above links to as well is well worth reading too. It takes on, among other things, the subject of meta-studies:

“Now let’s iterate this [...publication/attempted replication] process several times. Every couple of years, some enterprising young investigator will decide she’s going to try to replicate that cool effect from 2009, since no one else seems to have bothered to do it. This goes on for a while, with plenty of null results, until eventually, just by chance, someone gets lucky (if you can call a false positive lucky) and publishes a successful replication. And also, once in a blue moon, someone who gets a null result actually bothers to forces their graduate student to write it up, and successfully gets out a publication that very carefully explains that, no, Virginia, lawn gnomes don’t really make you happy. So, over time, a small literature on the hedonic effects of lawn gnomes accumulates.

Eventually, someone else comes across this small literature and notices that it contains “mixed findings”, with some studies finding an effect, and others finding no effect. So this special someone–let’s call them the Master of the Gnomes–decides to do a formal meta-analysis. (A meta-analysis is basically just a fancy way of taking a bunch of other people’s studies, throwing them in a blender, and pouring out the resulting soup into a publication of your very own.) Now you can see why the failure to publish null results is going to be problematic: What the Master of the Gnomes doesn’t know about, the Master of the Gnomes can’t publish about. So any resulting meta-analytic estimate of the association between lawn gnomes and subjective well-being is going to be biased in the positive directio. That is, there’s a good chance that the meta-analysis will end up saying lawn gnomes make people very happy,when in reality lawn gnomes only make people a little happy, or don’t make people happy at all.”

Some meta-analysts are more aware of the publication bias problem than others – I remember reading a meta-study by Martin Paldam a while ago where he emphasized this problem in the analysis, and I believe he’s actually done a meta-study on publication bias as well, though I don’t remember which subject it was about and I’m too lazy to look it up now. In some studies this issue is hardly even mentioned though.

If you read the links, you’ll become much better able to evaluate some of the stuff that’s out there. Btw. I am somewhat in agreement with Yvain when it comes to two main points: a) If it is indeed true that a lot of the stuff that gets published in medical journals later turn out to be wrong, the most likely explanation is that ‘the system’ is generally working and that we’re getting smarter over time, and b) the fact that the findings of ‘mainstream’ researchers are more prone to error that you might have thought does not make the non-mainstream people any less unlikely to be wrong.

Squick!

Don’t read the following passage if you’re prone to having nightmares about wasps paralyzing you, giving you a lobotomy and then laying eggs on your stomach, the larvae of which will eat you – starting from the outside, then slowly hollowing you out by going into your stomach and eating your internal organs one by one in a manner that will keep you alive as long as possible – while you’re buried alive in the burrow of the wasp (some people are weird, what do I know, maybe such nightmares are normal? …wait, did my warning just increase the likelihood that someone actually will have nightmares about this? Anyway, if you do have nightmares like those, you have an interesting mind!):

“As early as the 1940s it was reported that female wasps of this species [Ampulex compressa] sting a roach (specifically a Periplaneta americana, Periplaneta australasiae or Nauphoeta rhombifolia[1]) twice, delivering venom. A 2003 study[2] using radioactive labeling demonstrated that the wasp stings precisely into specific ganglia of the roach. It delivers an initial sting to a thoracic ganglion and injects venom to mildly and reversibly paralyze the front legs of its victim. This facilitates the second venomous sting at a carefully chosen spot in the roach’s head ganglia (brain), in the section that controls the escape reflex. As a result of this sting, the roach will first groom extensively, and then become sluggish and fail to show normal escape responses.[3] In 2007 it was reported that the venom of the wasp blocks receptors for the neurotransmitter octopamine.[4]

The wasp proceeds to chew off half of each of the roach’s antennae.[1] Researchers believe that the wasp chews off the antenna to replenish fluids or possibly to regulate the amount of venom because too much could kill and too little would let the victim recover before the larva has grown. The wasp, which is too small to carry the roach, then leads the victim to the wasp’s burrow, by pulling one of the roach’s antennae in a manner similar to a leash. Once they reach the burrow, the wasp lays a white egg, about 2 mm long, on the roach’s abdomen. It then exits and proceeds to fill in the burrow entrance with pebbles, more to keep other predators out than to keep the roach in.

With its escape reflex disabled, the stung roach will simply rest in the burrow as the wasp’s egg hatches after about three days. The hatched larva lives and feeds for 4–5 days on the roach, then chews its way into its abdomen and proceeds to live as an endoparasitoid. Over a period of eight days, the wasp larva consumes the roach’s internal organs in an order which guarantees that the roach will stay alive, at least until the larva enters the pupal stage and forms a cocoon inside the roach’s body. Eventually the fully grown wasp emerges from the roach’s body to begin its adult life. Development is faster in the warm season.

Adults live for several months. Mating takes about one minute, and only one mating is necessary for a female wasp to successfully parasitize several dozen roaches.

While a number of venomous animals paralyze prey as live food for their young, Ampulex compressa is different in that it initially leaves the roach mobile and modifies its behavior in a unique way. Several other species of the genus Ampulex show a similar behavior of preying on cockroaches.[1] The wasp’s predation appears only to affect the cockroach’s escape responses. Research has shown that while a stung roach exhibits drastically reduced survival instincts (such as swimming, or avoiding pain) for approximately 72 hours, motor abilities like flight or flipping over are unimpaired.[5][6]“

…

Here’s the article, here’s a direct link to the 2003 paper mentioned above.

An article on articles about articles

Here’s the link, this is great stuff and you really should read all of it. An excerpt:

“This is a news website article about a scientific paper

In the standfirst I will make a fairly obvious pun about the subject matter before posing an inane question I have no intention of really answering: is this an important scientific finding?”

In this paragraph I will state the main claim that the research makes, making appropriate use of “scare quotes” to ensure that it’s clear that I have no opinion about this research whatsoever.

In this paragraph I will briefly (because no paragraph should be more than one line) state which existing scientific ideas this new research “challenges”.

If the research is about a potential cure, or a solution to a problem, this paragraph will describe how it will raise hopes for a group of sufferers or victims.

This paragraph elaborates on the claim, adding weasel-words like “the scientists say” to shift responsibility for establishing the likely truth or accuracy of the research findings on to absolutely anybody else but me, the journalist.

In this paragraph I will state in which journal the research will be published. I won’t provide a link because either a) the concept of adding links to web pages is alien to the editors, b) I can’t be bothered, or c) the journal inexplicably set the embargo on the press release to expire before the paper was actually published.”

Growth of the human brain over time

Here’s the link, via Razib Khan.