Econstudentlog

Random stuff

i. I went to my first Mensa meeting a couple of days ago. It was nice, I felt very welcome. When I’m to participate in such a ‘new/unknown type of social event’, I always think a bit beforehand about how to approach things and how to handle various contingencies – below a couple of remarks I made in a skype conversation shortly before the meeting:

“I have been wondering what [would be] the best communications strategy today. I think I’ll probably just keep in the background, to the extent I’m allowed to do that, and observe what’s going on. It’s what I usually do when I’m in groups with lots of people. […]

On the other hand I should also use such activities to improve my social skills, and that involves actually talking to- and interacting with other people.. […]

And there’s a path-dependence aspect to consider as well. People usually categorize others and put them into some box quite quickly. If I’m put into the antisocial don’t-want-to-talk-with-others box from the start, it might be hard to break out of that later on.”

In the end I actually ended up talking myself into thinking that I should try to participate in conversations as much as possible, and try not to hold back like I usually do. As I put it in a later Skype conversation, “I actually made a point out of being significantly more verbose than I normally am. Maybe it was a bad strategic choice.”

Well, we’ll see. This was certainly not my last Mensa meeting.

ii. A look over my shoulder:

DSCN3698

DSCN3703

On a very slightly related point, I just started reading Tom Apostol’s Introduction to Analytic Number Theory.

iii. Another Stanford lecture:

You can easily skip the first 6 minutes without missing out on anything important. I think the first lecturer goes a bit overboard towards the end, and in general I’d say I preferred Erik Knudsen’s part of the lecture. I found it annoying that it was sometimes hard to figure out which figures or elements of a slide they were actually talking about – they’ll point to a specific part of a slide, but you can’t see where they’re pointing so you have no idea which dendrite ‘this dendrite’ actually is, and even though you can usually infer it it’s still confusing and suboptimal.

I know most readers don’t actually watch these lectures, so a couple of points to take with you that don’t require you to know much about the details:

“Human speech is about 500 to 3000 Hz. […] the low-frequency part of speech [are] vowels […] they’re lower tones […] Consonants […] give the sense to speech [and are higher frequency sounds]. And what happens as people lose high-frequency hearing [is] they become confused between the consonant sounds of speech. So hope and soap and cope begin to run together.” (I didn’t know this)

“the cues that you use to localise sounds in space are largely based on inter-oral differences in the timing and the level or intensity of sound in your two ears.”

They also argue (around the 1:30 mark) that given how important binaural interactions are, people shouldn’t wear hearing aids on just one ear.

iv. Schooling Is Not Education! – A Report of the Center for Global Development Study Group on Measuring Learning Outcomes. This point was also emphasized in one of my previous courses and I think I’ve covered this stuff before, but I haven’t linked to this paper yet. Some stuff from the paper:

“In India, national survey evidence reveals that only about one-third of children in grade 5 can perform long division, and one-third cannot perform two-digit subtraction.[11] Nearly one-half of grade 5 students cannot read a grade 2 text and one in five cannot follow a grade 1 text.[12] Sixty percent of Indian children enrolled in grade 8 cannot use a ruler to measure a pencil. Only 27 percent of Indian children who complete primary school can read a simple passage, perform division, tell time, and handle money, although students should master each of these skills by the end of the second year of school.[13] These statistics compare starkly with the official 81 percent youth literacy rate reported by the United Nations Education, Scientific and Cultural Organization (UNESCO).[14]

Similar findings have emerged elsewhere.[15] […]

A flat learning trajectory through successive school grades is reflected in low test scores among older students. Using several sources of recent data from India, the Center for Global Development’s Lant Pritchett examined the number of repeat questions that fourth, sixth, and eighth graders answered correctly. For language, the percentage climbs from 51 to 57 percent between fourth and eighth grades. For math, it climbs from 36 to 53 percent. This suggests that it would take 32 years of schooling for 90 percent of all students to correctly answer a language question that more than half of all fourth graders already correctly answered. India is hardly unique in its flat learning trajectories. Studies of the impact of education on learning in Bangladesh in the 1990s found that three additional years of schooling had no appreciable impact on learning achievement.[18]

At higher levels, results are perhaps even more worrying. Internationally comparable mathematics tests under the Trends in International Mathematics and Science Study (TIMSS) suggest that the average eighth grader in Ghana has a test score that would place her in the bottom 0.2 percent of US students. Even in considerably richer developing countries, the learning gap is large: the average Chilean student would be in the bottom 6.4 percent of US students, based on TIMSS scores. [19]”

Fig 3

Fig 4

“Until school systems can guarantee that students will learn while sitting in class, it may even be counterproductive to encourage longer periods of universal education.[24] In fact, expanded enrollments can actually harm overall learning outcomes if quality cannot be broadly maintained. While grade 8 enrollment in India increased from 82 to 87 percent from 2006 to 2011, ASER tests suggests the fraction of grade 8 children who could do division fell from 70 percent to 57 percent. This suggests that fewer school-age children actually learned division, despite climbing enrollments.” […]

“on an average school day, 11 percent of teachers are absent in Peru, 16 percent are absent in Bangladesh, and 27 percent are absent in Uganda.[29] Even when they are present, teachers may make limited efforts to create a friendly learning environment. ASER’s observation of rural education practices in 1,075 classrooms across five Indian states reveals that in only about a quarter of classrooms was a student witnessed asking a question. Other child-friendly practices […] were even less common.

Direct methods to improve teacher attendance and effort have improved test scores with mixed results,[31] as has the use of contract teachers.[32] The importance of systemic issues is demonstrated by the fact that interventions that successfully improve teacher performance in one system can fail when applied elsewhere. For example, Paul Atherton and Geeta Kingdon show that students taught by contract teachers in public schools in Uttar Pradesh learned about twice as much per year as those taught by civil service teachers.[33] At the same time, an attempt to scale contract teachers in Kenya revealed that contract teachers only influenced test scores when hired by a nongovernmental organization, not the Kenyan government.[34] A randomized study in the Indian state of Andhra Pradesh suggests that bonuses for teachers based on exam results can improve outcomes.[35] In Kenya, however, teacher performance incentives related to test scores increased student learning only in the short run.[36] […]

“It is perhaps not surprising that there is a gap between schooling and learning, and that education reform is difficult. Schools and education systems are about a lot more than learning. For students, schools are also about signaling innate intelligence, status, and social networks. For parents, they are also a form of daycare. For teachers, they are a stable source of income. For governments, schools are also about socialization, employment, and rent generation.[45] A complex story of political economy lies behind the schooling-learning gap. Given the complex political economy of systemic reform, and the considerable diversity in existing educational systems around the world, solutions to learning stagnation will vary immensely across countries.”

Fig 5

September 12, 2013 Posted by | Books, Economics, education, Lectures, Mathematics, Medicine, Personal | Leave a comment

Life expectancy gaps related to income/education

I’m currently writing a topic on ‘the causal effect of education on health’, so this is a topic I’ve looked at a bit – consider this post a ‘workblog’-post, even though it’s only tangentially related to what I’m working on.

This kind of stuff – health disparities related to education and income – pops up in the public debate every now and then, see e.g. this recent article (in Danish), or this analysis by AE-rådet (also in Danish). This is ‘politics’ to some extent (see the previous post), but it’s also a question about what’s actually going on in the world, and the latter type of question is the type of question I tend to be interested in answering. I’d like to make some general points here which are sometimes overlooked:

i. People with lower education are fatter. And being fat is bad for your health.

ii. People with lower levels of education smoke more: “Well-documented declines in smoking prevalence over time have not occurred evenly throughout society (12, 13). They have been most substantial among the most educated. Thus, the least educated form increasing proportions of those who remain smokers.” Regarding alcohol the picture is more complicated (as I’ve talked about before), however it should be noted that if the variance of the quantity consumed by the highly educated is lower than for the lower educated groups, as they claim in the article I link to at the beginning of this paragraph, then it would make sense if the highly educated people who die from alcohol-related diseases die later and lose fewer years of their life to the alcohol than does the group with low education (‘the uneducated alcoholic loses 20 years, the educated alcoholic loses five…’). Either way alcohol matters much less than smoking, and the differences aren’t that big in the former case. Incidentally the causal pathways of the smoking link are still unclear: “The causal pathways between education and smoking are both complicated and contested in the literature.” (link)

iii. Lifestyle differences among different educational groups make up a big part of the difference in health outcomes: “the mediating effects of health behaviors – measured by smoking, drinking, exercising and the body mass index – account in the short run for 17% to 31% and in the long run for 23% to 45% of the entire effect of education on health, depending on gender.”

iv. An additional point related to point iii.: I haven’t looked for studies on this because it’s obvious, but the health gradient is more sensitive to stuff like income level and employment status in countries like the US than it is in Denmark. So international (non-Scandinavian?) estimates of the magnitude of educational effects and income effects on health outcomes are likely to be biased upwards, compared to what the magnitude would be in a country like Denmark where ability to pay for medical services problems are unlikely to have much influence on life expectancy at this point.

v. I’ll spell out this point even though it should be obvious by now: Many of the reasons why people with a low education on average die too soon relate to the fact that they on average make poorer choices when it comes to their health. And the stuff mentioned above is just a small part of what’s going on; you also have related stuff like information channels and compliance differences, on top of stuff like ‘likelihood of seeking proper medical attention conditional on you actually needing it, and ability to verbalize complaints so that the doctor makes the correct inferences’ (e.g. a lot of T2 diabetics don’t get diagnosed, and this lowers their life expectancy significantly).

vi. Note that whereas it’s true that some jobs are still more unhealthy than others (a traditional mechanism most people think of when they’re thinking about these things), if the connection between type of work and health risks is known people employed in such jobs would be expected to earn a risk premium – this is not super relevant when you look at education and health, but it is something to have in mind when analyzing health and income stuff.

vii. It should be noted that if you get better over time at treating people for stuff that isn’t lifestyle-related and so stop a lot of people from dying early on of other causes, then lifestyle-stuff is going to become a big driver of health disparities.

April 18, 2013 Posted by | education, health, Papers | 1 Comment

Richard Feynman – The Character of Physical Law

Did I ever blog any of these at some point? I’ve seen a couple of them before so I’m not sure, but I couldn’t find them in the archives and so I decided to post them here. I spent yesterday evening in great company:

They’ve held up quite well I think, and he made me smile/laugh multiple times each lecture. The last lecture had me looking around a bit on wikipedia.

If you want some newer physics lectures, Muller’s lectures (start here) are great too. I haven’t looked at this stuff yet, but that’s another source to consider if you’re curious about these things.

November 17, 2012 Posted by | education, Lectures, Physics | 4 Comments

Stuff

i. Temporal view of the costs and benefits of self-deception, by Chance, Nortona, Ginob, and Ariely. The abstract:

“Researchers have documented many cases in which individuals rationalize their regrettable actions. Four experiments examine situations in which people go beyond merely explaining away their misconduct to actively deceiving themselves. We find that those who exploit opportunities to cheat on tests are likely to engage in self-deception, inferring that their elevated performance is a sign of intelligence. This short-term psychological benefit of self-deception, however, can come with longer-term costs: when predicting future performance, participants expect to perform equally well—a lack of awareness that persists even when these inflated expectations prove costly. We show that although people expect to cheat, they do not foresee self-deception, and that factors that reinforce the benefits of cheating enhance self-deception. More broadly, the findings of these experiments offer evidence that debates about the relative costs and benefits of self-deception are informed by adopting a temporal view that assesses the cumulative impact of self-deception over time.”

A bit more from the paper:

“People often rationalize their questionable behavior in an effort to maintain a positive view of themselves. We show that, beyond merely sweeping transgressions under the psychological rug, people can use the positive outcomes resulting from negative behavior to enhance their opinions of themselves—a mistake that can prove costly in the long run. We capture this form of self-deception in a series of laboratory experiments in which we give some people the opportunity to perform well on an initial test by allowing them access to the answers. We then examine whether the participants accurately attribute their inflated scores to having seen the answers, or whether they deceive themselves into believing that their high scores reflect new-found intelligence, and therefore expect to perform similarly well on future tests without the answer key.

Previous theorists have modeled self-deception after interpersonal deception, proposing that self-deception—one part of the self deceiving another part of the self—evolved in the service of deceiving others, since a lie can be harder to detect if the liar believes it to be true (1, 2). This interpersonal account reflects the calculated nature of lying; the liar is assumed to balance the immediate advantages of deceit against the risk of subsequent exposure. For example, people frequently lie in matchmaking contexts by exaggerating their own physical attributes, and though such deception might initially prove beneficial in convincing an attractive prospect to meet for coffee, the ensuing disenchantment during that rendezvous demonstrates the risks (3, 4). Thus, the benefits of deceiving others (e.g., getting a date, getting a job) often accrue in the short term, and the costs of deception (e.g., rejection, punishment) accrue over time.

The relative costs and benefits of self-deception, however, are less clear, and have spurred a theoretical debate across disciplines (5–10). […]

As we had expected, social recognition exacerbated self-deception: those who were commended for their answers-aided performance were even more likely to inflate their beliefs about their subsequent performance. The fact that social recognition, which so often accompanies self-deception in the real world, enhances self-deception has troubling implications for the prevalence and magnitude of self-deception in everyday life.”

ii. Nonverbal Communication, by Albert Mehrabian. Some time ago I decided that I wanted to know more about this stuff, but I haven’t really gotten around to it until now. It’s old stuff, but it’s quite interesting. Some quotes:

“The work of Condon and Ogston (1966, 1967) has dealt with the synchronous relations of a speaker’s verbal cues to his own and his addressee’s nonverbal behaviors. One implication of their work is the existence of a kind of coactive regulation of communicator-addressee behaviors which is an intrinsic part of social interaction and which is certainly not exhausted through a consideration of speech alone. Kendon (1967a) recognized these and other functions that are also served by implicit behaviors, particularly eye contact. He noted that looking at another person helps in getting information about how that person is behaving (that is, to monitor), in regulating the initiation and termination of speech, and in conveying emotionality or intimacy. With regard to the regulatory function, Kendon’s (1967a) findings showed that when the speaker and his listener are baout to change roles, the speaker looks in the direction of his listener as he stops talking, and his listener in turn looks away as he starts speaking. Further, when speech is fluent, the speaker looks more in the direction of his listener than when his speech is disrupted with errors and hesitations. Looking away during these awkward moments implies recognition by the speaker that he has less to say, and is demanding less attention from his listener. It also provides the speaker with some relief to organize his thoughts.

The concept of regulation has also been studied by Scheflen (1964, 1965). According to him, a communicator may use changes in posture, eye contact, or position to indicate that (1) he is about to make a new point, (2) he is assuming an attitude relative to several points being made by himself or his addresse, or (3) he wishes to temporarily remove himself from the communication situation, as would be the case if he were to select a great distance from the addressee or begin to turn his back on him. There are many interesting aspects of this regulative function of nonverbal cues that have been dealt with only informally. […]

One of the first attempts for a more general characterization of the referents of implicit behavior and, therefore, possibly of the behaviors themselves, was made by Schlosberg (1954). He suggested a three-dimensional framework involving pleasantness-unpleasantness, sleep-tension, and attention-rejection. Any feeling could be assigned a value on each of these three dimensions, and different feelings would correspond to different points in this three-dimensional space. This shift away from the study of isolated feelings and their corresponding nonverbal cues and toward a characterization of the general referents of nonverbal behavior on a limited set of dimensions was seen as beneficial. It was hoped that it could aid in the identification of large classes of interrelated nonverbal behaviors.

Recent factor-analytic work by Williams and Sundene (1965) and Osgood (1966) provided further impetus for characterizing the referents of implicit behavior in terms of a limited set of dimensions. Williams and Sundene (1965) found that facial, vocal, or facial-vocal cues can be categorized primarily in terms of three orthogonal factors: general evalution, social control, and activity.

For facial expression of emotion, Osgood (1966) suggested the following dimensions as primary referents: pleasantness (joy and glee versus dread and anxiety), control (annoyance, disgust, contempt, scorn, and loathing versus dismay, bewilderment, surprise, amazement, and excitement), and activation (sullen anger, rage, disgust, scorn, and loathing versus despair, pity, dreamy sadness, boredom, quiet pleasure, complacency, and adoration). […]

Scheflen (1964, 1965, 1966) provided detailed observations of an informal quality on the significance of postures and positions in interpersonal situations. Along similar lines, Kendon (1967a) and Exline and his colleagues explored the many-faceted significance of eye contact with, or observation of, another […] These investigations consistently found, among same-sexed pairs of communicators, that females generally had more eye contact with each other than did males; also, members of both sexes had less eye contact with one another when the interaction between them was aversive […] In generally positive exchanges, males had a tendency to decrease their eye contact over a period of time, whereas females tended to increase it (Exline and Winters, 1965). […]

extensive data provided by Kendon (1967a) showed that observation of another person duing a social exchange varied from about 30 per cent of 70 per cent, and that corresponding figures for eye contact ranged from 10 per cent to 40 per cent. […]

Physical proximity, touching, eye contact, a forward lean rather than a reclining position, and an orientation of the torso toward rather than away from an addressee have all been found to communicate a more positive attitude toward him. A second set of cues that indicates postural relaxation includes asymmetrical placement of the limbs, a sideways lean and/or reclining position by the seated communicator, and specific relaxation measures of the hands or neck. This second set of cues relates primarily to status differences between the communicator and his addressee: there is more relaxation with an addressee of lower status, and less relaxation with one of higher status. […]

In sum, the findings from studies of posture and position and subtle variations in verbal statements […] show that immediacy cues primarily denote evaluation, and postural relaxation ues denote status or potency in a relationship. It is interesting to note a weaker effect: less relaxation of one’s posture also conveys a more positive attitude toward another. One way to interpret this overlap of the referential significance of less relaxation and more immediacy in communicating a more positive feeling is in terms of the implied positive connotations of higher status in our culture. A respectful attitude (that is, when one conveys that the other is of higher status) does indeed have implied positive connotations. Therefore it is not surprising that the communication of respect and of positive attitude exhibits some similarity in the nonverbal cues that they require. However, whereas the communication of liking is more heavily weighted by variations in immediacy, that of respect is weighted more by variations in relaxation.”

I should probably note here that whereas it makes a lot of sense to be skeptical of some of the reported findings in the book, simply to get an awareness of some of the key variables and some proposed dynamics may actually be helpful. I don’t know how deficient I am in these areas because I haven’t really given body language and similar stuff much thought; I assume most people haven’t/don’t, but I may be mistaken.

iii. A friend let me know about this ressource and I thought I should share it here. It’s a collection of free online courses/lectures provided by Yale University.

iv. Prevalence, Heritability, and Prospective Risk Factors for Anorexia Nervosa. It’s a pretty neat setup: “During a 4-year period ending in 2002, all living, contactable, interviewable, and consenting twins in the Swedish Twin Registry (N = 31 406) born between January 1, 1935, and December 31, 1958, underwent screening for a range of disorders, including AN. Information collected systematically in 1972 to 1973, before the onset of AN, was used to examine prospective risk factors for AN.”

Results  The overall prevalence of AN was 1.20% and 0.29% for female and male participants, respectively. The prevalence of AN in both sexes was greater among those born after 1945. Individuals with lifetime AN reported lower body mass index, greater physical activity, and better health satisfaction than those without lifetime AN. […]

[…]

This study represents, to our knowledge, the largest twin study conducted to date of individuals with rigorously diagnosed AN. Our results confirm and extend the findings of previous studies on prevalence, risk factors, and heritability.

Consistent with several studies, the lifetime prevalence of AN identified by all sources was 1.20% in female participants and 0.29% in male participants, reflecting the typically observed disproportionate sex ratio. Similarly, our data show a clear increase in prevalence of DSM-IV AN (broadly and narrowly defined) with historical time in Swedish twins. The increase was apparent for both sexes. Hoek and van Hoeken3 also reported a consistent increase in prevalence, with a leveling out of the trajectory around the 1970s. Future studies in younger STR participants will allow verification of this observation.

Several observed differences between individuals with and without AN were expected, ie, more frequent endorsement of symptoms of eating disorders. Other differences are noteworthy. Consistent with previous observations, individuals with lifetime AN reported lower BMIs at the time of interview than did individuals with no history of AN. Although this could be partially accounted for by the presence of currently symptomatic individuals in the sample, our results remained unchanged when we excluded individuals likely to have current AN (ie, current BMI, ≤17.5). Previous studies have shown that, even after recovery, individuals with a history of AN have a low BMI.59 Although perhaps obvious, a history of AN appears to offer protection against becoming overweight. The protective effect also holds for obesity (BMI, ≥30), although there were too few individuals in the sample with histories of AN who had become obese for meaningful analyses. Despite the obvious nature of this observation, the mechanism whereby protection against overweight is afforded is not immediately clear. Those with a history of AN reported greater current exercise and a perception of being in better physical health. One possible interpretation of this pattern of findings is that individuals with a history of AN continue to display subthreshold symptoms of AN (ie, excessive exercise and caloric restriction) that contribute to their low BMIs. Alternatively, symptoms that were pathologic during acute phases of AN, such as excessive exercise and decreased caloric intake, may resolve over time into healthy behaviors, such as consistent exercise patterns and a healthful diet, that result in better weight control and self-rated health.

Regardless of which of these hypotheses is true, another intriguing difference is that individuals with lifetime AN report a lower age at highest BMI, although the magnitude of the highest lifetime BMI does not differ in those with and without a history of AN. Those with AN report their highest lifetime BMIs early in their fourth decade of life on average, whereas those without AN report their highest BMIs in the middle of their fifth decade of life (close to the age at interview). On a population level, adults tend to gain on average 2.25 kg (5 lb) per decade until reaching their eighth decade of life.60 Although more detailed data are necessary to make definitive statements about different weight trajectories, our results suggest not only that individuals with AN may maintain low BMIs but also that they may not follow the typical adult weight gain trajectories. These data are particularly intriguing in light of recent reports of AN being associated with reduced risk of certain cancers61 – 62 and protective against mortality due to diseases of the circulatory system.63 – 64 Energy intake is closely related to fat intake and obesity, both of which have also been related to cancer development65 – 66 and both of which are reduced in AN. Further detailed studies of the weight trajectories and health of individuals with histories of AN are required to explicate the nature and magnitude of these intriguing findings.

Of the variables assessed in 1972 to 1973, neuroticism emerged as the only significant prospective predictor of AN. This is notable because there have been few truly prospective risk factor studies of AN.”

v. The music is a bit much for me towards the end, but this is just an awesome video. I think I’d really have liked to know that guy:

vi. Political Sorting in Social Relationships: Evidence from an Online Dating Community, by Huber and Malhotra.

I found these data surprising (and I’m skeptical about the latter finding):

“Among paid content, online dating is the third largest driver of Internet traffic behind music and games (Jupiter Research 2011).A substantial number of marriages also result from interactions started online. For instance, a Harris Interactive study conducted in 2007 found that 2% of U.S. marriages could be traced back to relationships formed on eHarmony.com, a single online dating site (Bialik 2009).”

Anyway I’ll just post some data/results below and leave out the discussion (click to view tables in full size). Note that there are a lot of significant results here:

The last few figures are also interesting (people really care about that black/white thing when they date (online)…). but you can go have a look for yourself. As I’ve already mentioned there are a lot of significant results – they had a huge number of data to work with (170,413 men and 132,081 women).

vii. John Nash on Cryptography.

November 16, 2012 Posted by | Books, Cryptography, Data, dating, education, Papers, Psychology, Random stuff | Leave a comment

Danish education – some numbers

The data included in this post are from Statistics Denmark, Statistikbanken – “KRHFU1: Befolkningens højeste fuldførte uddannelse (15-69 år) efter område, herkomst, uddannelse alder og køn.” I had a look at the documentation and most of the data are registry data, but the data on immigrants are survey-based (and thus less reliable) – no surveys have been conducted since 2006, so all immigrants who’ve arrived since then have an ‘unknown education level’ in the data. If you’re more curious about that subject I presented some other, much more detailed, data on the same topic a while ago here. If you disaggregate the data on immigrants and descendants, the image looks worse than it does here because descendants of Western immigrants have a different age profile than do descendants of non-Western immigrants – one third of descendants of Western immigrants are above the age of 30, whereas only 6% of non-Western descendants are that old (also, only 10% of all descendants in Denmark have reached the age of 30). Another aspect adding to the confusion is the fact that Western immigrants are quite well educated, which contributes to the confusion about the numbers – if you look at ‘Danish immigrants’, you’re basically mixing data drawn from two completely different distributions.

If you want to know more about the Danish education system this link may be of some use (note that there’s a lot of additional stuff in the sidebar there). Regarding the higher education stuff, a short-cycle higher education is at most two years long, a medium-cycle higher education is from 2-4 years long. The latter differs from a ‘standard’ Danish bachelor’s degree: “Professionally oriented higher education programmes are offered at colleges. Whereas in other countries, similar programmes may be offered by universities, in Denmark they have traditionally been offered by specialised colleges” (from the link). Long-cycle higher education corresponds to a Master’s degree. Click to view graphs in full size. All data reported are from the 2012 data sets. There aren’t all that many data included in this post, but part of the reason is that the source only gave the raw data – not percentage stuff (which is much more informative) – and variable transformations take time. Anyway…

So, let’s have a look at the data… I figured it’d be interesting to look at the first cohort of ‘young people’ (30-34) consisting of people about whom we can say with relative certainty that almost all of them have completed their formal education. I’ll start out with the males:

Descendant n is not that high compared to the other groups, but I think it’s ‘high enough to draw conclusions’ from the data (n=3041). Descendants are more than twice as likely to not get any education aside from grundskole than are people of Danish origin. What about the females?

Females are more likely to get an education so the numbers generally look better. When including all groups, 17,7% of males and 12,9% of females end up with only grundskole – it’s a significant difference, but it’s not actually that high compared to some of the variation we see in these data. The female descendants (n=2873) are roughly twice as likely to only have a grundskole education (22.8% vs 11,9%) as are people of Danish origin. Male immigrants were much less likely to have attended vocational school than were males of Danish origin; that difference is almost gone when we’re looking at the females. As I’ve mentioned elsewhere, do have in mind that a lot of those ‘unknown education level’ immigrants are people with very little education, and/or education which is not worth a lot on the Danish labour market – immigrants who’ve studied here have known education levels, and most of the people with an unknown education level aren’t highly educated foreigners who love the Danish weather and our marginal tax rates.

Education rates have increased over time. Below I’ve compared the numbers for males at the age of 65-69 with the 30-34 year old cohort. I didn’t really see why I should care about the education levels of immigrants or descendants in that age group, so I only included people of Danish origin:

Note here that the mandatory education level back then was lower than it is now (7 years vs 9 years), so most of the people in this graph with only grundskole education have spent less time in school than have the groups in the sections above. The numbers are not identical, but they’re not that different given how much the educational system changed over that 35 year period. I think it’s interesting that ‘only high school’ (or technical/trade high school) was a less likely scenario for people in this group than for the younger generation, but on the other hand it’s not surprising.

How about the females?

The difference is significant. The number of uneducated women has been reduced dramatically and the number of females who are highly educated has gone up a lot. The proportion of 30-34 year old females of Danish origin with a long-cycle higher education is higher than the proportion of males of Danish origin with a long-cycle higher education (15.6% vs 13.2%). The same pattern is seen in the younger 25-29 year old cohort: In that age group, 6,86% of males of Danish origin (n=8791) have completed a long-cycle higher education, whereas the corresponding number for females of Danish origin (n=10168) is 8,22% (which is a 20% difference). Here’s a mapping of all relevant cohorts included in the data set:

It’s well known that females are more likely to get a medium-cycle higher education than are males (n=34921 or 25,6% of females and n=14750 or 10,6% of males at the ages of 30-34 have such an education), so I decided to also look at the proportion of the genders with any type of higher education (short-cycle, medium-cycle, bachelor, long-cycle or PhD) and condition on age. A lot of people would probably be surprised to learn that it looks this way:

In case you want to argue that the short-cycle ones are roughly equivalent to vocational schooling (‘females take short-cycle higher educations where males take vocational schooling’), it’s worth noting that for all age groups males are more likely to get a short-cycle higher education than are females. No, the difference derives from the other categories. It seems that females are better educated than males on average and have been for decades – I did not expect that.

The dataset also includes some information about geographical variation. The percentage of people with a long-cycle higher education varies a lot:

(Wikipedia can help you if you don’t know anything about the Danish regions). If you meet a random person above the age of 30 on the street, he or she is more than three times as likely to have a long-cycle higher education if the person is from Copenhagen than if the person is from the Region of Southern Denmark. If you restrict your search to include only people drawn from the younger cohorts, the differences become even larger. If you scale down further and look at the differences at the municipal level, they are just huge; to take but one example, 3,1% of Danes above 30 years of age from the Jammerbugt municipality have a long-cycle higher education, whereas the corresponding number for people living in the Gentofte municipality is 29,6%.

November 8, 2012 Posted by | Data, Demographics, education | 2 Comments

Stuff

i. From The future of infectious diseases. Watch it.:

“Take home messages:

  • You can watch evolution in progress in quickly-reproducing organisms, like malaria
  • Over 100,000 Americans die of infections that were easily treated 30 years ago due to the evolution of resistance (twice the number of people who will die in car crashes).
  • In an arms race between us and infectious diseases, we lose.
  • We need to understand the evolutionary forces unleashed by medicine before we can manage infectious disease
  • We need to ask, “Will (this drug) STAY safe, and CONTINUE to work”, not just if it is safe and whether it will work.
  • The Lancet (a high impact medical journal) rejected an evolutionary paper addressing malaria because, “a good understanding of evolutionary biology is beyond most of our readers.””

Point 3 is one I try to remember to bring up every time I find myself in a discussion about matters related to how the future development of medicine will look like. Unless you do not agree with that one, it’s very hard to be an optimist about the future of medicine.

ii. I discussed this subject briefly yesterday, and I later started thinking about whether I’d actually blogged this (pdf) publication (in Danish, sorry), PISA København 2010, which deals with the educational achievements of Danish children in Copenhagen who left the 9th grade in 2010. I don’t think I have (I couldn’t find anything in the archives), so I decided to add a link here as well as a few observations from the paper:

“Opdeles eleverne efter andelen af indvandrerelever på deres skoles niende klassetrin, finder man generelt, at jo større andel af indvandrere, des lavere gennemsnitlig læsetestscore.” (a (loose) translation: ‘If the students are distributed according to the proportion of immigrant-pupils in the 9th grade, a general finding is that the larger the share of immigrants, the lower the average reading test score’).

Immigrant groups perform worse than non-immigrant groups, and the proportion of immigrants also affects the performance of the non-immgrant pupils (negatively) for some, though not all, specifications. The ‘Danish’ pupils enrolled in schools where the proportion of immigrant pupils exceeds 50% do significantly worse than do Danish pupils who are enrolled in schools where the immigrant pupil proportion is below 25% (p.11). Immigrant pupils also do better in schools with less than 25% immigrants than they do in schools where the proportion of immigrant pupils exceeds that number (p.11).

A table from the report (p.31), click to view full size:

The above table contains some numbers related to PISA’s reading test, with a special focus on the proportion of pupils in the sample who are functionally illiterate, corresponding to a reading performance of less than level 2 on the PISA scale (which is described in more details in the Appendix, p. 82-83 – I will not go into details here unless asked). In 2010, 14% of ‘Danish’ pupils and 42% of ‘immigrant pupils’ from schools in Copenhagen were functionally illiterate judging from the PISA reading test. There’s a big gender gap – 17% of the girls and 30% of the boys were functionally illiterate. The difference between the performances of first (44%) and second (41%) generation immigrant pupils is not statistically significant. Almost half of all 9th grade immigrant pupils in the public school system – 48% of first generation immigrant pupils in public schools and 46% of second generation immigrant pupils in public schools – were functionally illiterate.

There’s a lot of hidden variation in the immigrant numbers and not all immigrant groups do equally badly. It’s worth having in mind that these results are actually averages. Taking not-insignificant heterogeneity in the immigrant sample into account, it’s surely the case that some immigrant groups do even worse than these numbers might imply. If you look at the school level, some of the numbers probably get much worse. The Rockwool Foundation found in 2007 that 64% of pupils of Arab origin in the 9th grade were functionally illiterate (Danish link). In the PISA report they don’t go into much details, but they do note that pupils of Lebanese (/Palestinians), Iraqi and Turkish origin do worse than do pupils of Pakistani origin (also from p. 11).

iii. Another paper, Moral Hypocrisy, Power and Social Preferences, by Rustichini and Villeval (via Robin Hanson):

“Abstract: We show with a laboratory experiment that individuals adjust their moral principles to the situation and to their actions, just as much as they adjust their actions to their principles. We first elicit the individuals’ principles regarding the fairness and unfairness of allocations in three different scenarios (a Dictator game, an Ultimatum game, and a Trust game). One week later, the same individuals are invited to play those same games with monetary compensation. Finally in the same session we elicit again their principles regarding the fairness and unfairness of allocations in the same three scenarios.

Our results show that individuals adjust abstract norms to fit the game, their role and the choices they made. First, norms that appear abstract and universal take into account the bargaining power of the two sides. The strong side bends the norm in its favor and the weak side agrees: Stated fairness is a compromise with power. Second, in most situations, individuals adjust the range of fair shares after playing the game for real money compared with their initial statement. Third, the discrepancy between hypothetical and real behavior is larger in games where real choices have no strategic consequence (Dictator game and second mover in Trust game) than in those where they do (Ultimatum game). Finally the adjustment of principles to actions is mainly the fact of individuals who behave more selfishly and who have a stronger bargaining power.
The moral hypocrisy displayed (measured by the discrepancy between statements and actions chosen followed by an adjustment of principles to actions) appears produced by the attempt, not necessarily conscious, to strike a balance between self-image and immediate convenience.”

iv. False-Positive Psychology: Undisclosed Flexibility in Data Collection and Analysis Allows Presenting Anything as Significant, by Simmons, Nelson and Simonsohn. A pretty neat paper:

“Abstract
In this article, we accomplish two things. First, we show that despite empirical psychologists’ nominal endorsement of a low rate of false-positive findings (! .05), flexibility in data collection, analysis, and reporting dramatically increases actual false-positive rates. In many cases, a researcher is more likely to falsely find evidence that an effect exists than to correctly find evidence that it does not. We present computer simulations and a pair of actual experiments that demonstrate how unacceptably easy it is to accumulate (and report) statistically significant evidence for a false hypothesis. Second, we suggest a simple, low-cost, and straightforwardly effective disclosure-based solution to this problem. The solution involves six concrete requirements for authors and four guidelines for reviewers, all of which impose a minimal burden on the publication process.”

v. Cognitive Sophistication Does Not Attenuate the Bias Blind Spot, by West, Meserve & Stanovich:

“The so-called bias blind spot arises when people report that thinking biases are more prevalent in others than in themselves. Bias turns out to be relatively easy to recognize in the behaviors of others, but often difficult to detect in one’s own judgments. Most previous research on the bias blind spot has focused on bias in the social domain. In 2 studies, we found replicable bias blind spots with respect to many of the classic cognitive biases studied in the heuristics and biases literature (e.g., Tversky & Kahneman, 1974). Further, we found that none of these bias blind spots were attenuated by measures of cognitive sophistication such as cognitive ability or thinking dispositions related to bias. If anything, a larger bias blind spot was associated with higher cognitive ability. Additional analyses indicated that being free of the bias blind spot does not help a person avoid the actual classic cognitive biases. We discuss these findings in terms of a generic dual-process theory of cognition.”

I’ll just repeat part of that abstract: “none of these bias blind spots were attenuated by measures of cognitive sophistication such as cognitive ability or thinking dispositions related to bias. If anything, a larger bias blind spot was associated with higher cognitive ability.

A few other remarks from the paper (but do read all of it if you find the result interesting):

“the bias blind spot joins a small group of other effects such as myside bias and noncausal base-rate neglect (Stanovich & West, 2008b; Toplak & Stanovich, 2003) in being unmitigated by increases in intelligence. That cognitive sophistication does not mitigate the bias blind spot is consistent with the idea that the mechanisms that cause the bias are quite fundamental and not easily controlled strategically — that they reflect what is termed Type 1 processing in dual-process theory (Evans, 2008; Evans & Stanovich, in press). Two of the theoretical explanations of the effect considered by Pronin (2007)—naive realism and defaulting to introspection—posit the bias as emanating from cognitive mechanisms that are evolutionarily and computationally basic. Much research on the bias blind spot describes the asymmetry in bias detection in self compared to others as being spawned by a belief in naive realism—the idea that one’s perception of the world is objective and thus would be mirrored by others who are open-minded and unbiased in their views (Griffin & Ross, 1991; Pronin et al., 2002; Ross & Ward, 1996). Naive realism is developmentally primitive (Forguson & Gopnik, 1988; Gabennesch, 1990) and thus likely to be ubiquitous and operative in much of our basic information processing.

[rereading this, it reminded me of this quote, from a recent lesswrong article: “if you aren’t treating humans more like animals than most people are, then you’re modeling humans poorly. You are not an agenty homunculus “corrupted” by heuristics and biases. You just are heuristics and biases. And you respond to reinforcement, because most of your motivation systems still work like the motivation systems of other animals.”]

It is likewise with self-assessment based on introspective information, rather than behavioral information (Pronin & Kugler, 2007). The bias blind spot arises, on this view, because we rely on behavioral information for evaluations of others, but on introspection for evaluations of ourselves. The biases of others are easily detected in their overt behaviors, but when we introspect we will largely fail to detect the unconscious processes that are the sources of our own biases (Ehrlinger et al., 2005; Kahneman, 2011; Pronin et al., 2004; Wilson, 2002). When we fail to detect evidence of bias, we are apt to decide no bias has occurred and that our decision-making process was indeed objective and reasonable. This asymmetry in bias assessment information has as its source a ubiquitous and pervasive processing tendency— introspective reliance — that again is developmentally basic (Dennett, 1991; Sterelny, 2003).”

vi. Via Gwern, a meta-analysis on depression and exercise. There seems to be a short-term positive effect, but “there is little evidence of a long-term beneficial effect of exercise in patients with clinical depression.”

June 22, 2012 Posted by | Data, Demographics, education, Infectious disease, Papers, Pharmacology, Psychology | 2 Comments

Education at a glance 2009: OECD Indicators

I thought about just giving you the link, but I guess I should probably post a bit of data here as well. There’s a lot more stuff at the link. Please refrain from asking me questions about the assumptions underlying these estimates in the comments; if you care enough to ask about that kind of stuff, then you also care enough to click the link and figure out what’s going on yourself.

April 11, 2012 Posted by | Data, Demographics, education | Leave a comment

Having fun

(Click to view full size:)

I spent most of the day doing exercises. 10 hours or so. Then an hour’s worth of reading on the side. I think perhaps I’d have found this stuff interesting 4-5 years ago.

Imagine how much fun it is to spend your Saturday doing this stuff while feeling guilty about not doing even more of it, even though you pretty much hate every second of your life you spend on it, all the while feeling that it’s futile anyway because you’ll probably just fail.

The funny thing is that if you add the total number of hours I’ve spent on this course (combined, remember that I’m retaking it this semester), I doubt anyone who got less than an A would be even close to that total time expenditure. I’ll consider myself very lucky if I get a C. I think he failed something like one-fourth/one-third of the class at the original exam in January.

Don’t expect answers to this post, I’ve been offline all day and I’m not sure I’ll go online again before the exam.

February 18, 2012 Posted by | Economics, education, Personal | Leave a comment

Data on Danish immigrants, 2011 (3)

The third post in the series, here are the first two posts. This part will deal with education and I must admit that it’s less data-heavy than the previous two posts, in part because I felt it was necessary to spend some time explaining how the Danish education system actually works here (and in part because I feel there’s a limit as to how much time I can justify spending on posts like these). I’ll do another post on crime later on, so this is not the last post in the series. Anyway, here goes:

*In 2010, 44% of male descendants of non-Western immigrants and 61% of female descendants of non-Western immigrants in Denmark at the age of 30 had finished an education leading to a vocational/professional qualification (see below for some notes on terminology). The corresponding numbers for people of Danish origin at the age of 30 were 73% and 79%. The education level of non-Western female descendants has increased over time; in 2004 the number was 44%. (p.65)

*It was a bit harder to translate stuff from this section than the rest because the Danish education system is a bit different from that of e.g. the US, creating a few problems related to terminology. The terminology I’ve used in this section when I was in doubt follows this source. So, which educations are in fact included in the ‘education leading to a …’ (abbreviated ELVQs in the following) measure above and which are not? ELVQs include (Danish link) various technical educations (electrician, carpenter,…), further education leading to a degree (BA, MA, PhD) as well as various other educations (office education, teaching, nursing,…). A high school degree is not included in the set, nor is a grundskoleuddannelse (see below), and if you’re a college drop-out who have not obtained a degree you’re also not included in the set of people with an ELVQ. The idea is of course that if you have an ELVQ, you have finished an education that has given you some specific skills that are useful in terms of finding and retaining employment. I decided this would also be as good a place as any to add a bit more background info about the Danish education system you might need to make sense of the numbers in the report – it’s not in there, so no page references. In Denmark the lowest attainable ‘formal education level’ (i.e. disregarding drop-outs before that point) you can have is completion of the 9th grade (grundskoleuddannelse). The graduation exam is called ‘Folkeskolens afgangsprøve’. Technically it’s a little complicated as to where exactly to put high school in terms of grades, because some people finish 9th grade and then go to high school directly (I did) whereas others take 10th grade first at the same place they took 1st-9th grade before they go to high school. The coursework in Danish high schools is the same for people who went to 10th grade before going to HS and for people who didn’t, and HS classes are a mix of both types of students. I’m not completely sure if you’re required to take 10th grade before you can enroll in a vocational(/technical) education like carpentry, but I think some of them do demand that you have 10th grade before you can start, or at least that you have taken some of the specific courses (Danish, maths). Adult immigrants without an education can take a ‘basic adult education’ which is supposed to confer the same skills as a traditional grundskoleuddannelse (in a shorter amount of time) – after they have that they can move on to a vocational education or secondary education.

*A Danish ELVQ perhaps needless to say significantly increases employment opportunities. For 30-39 year old male non-Western immigrants who had only a grundskoleuddannelse/basic adult education, the employment rate was 58% in 2010 (females: 45%, p.79). For those with a vocational education, the employment rate was 76% (females: 78%). For those with a medium-cycle higher education (‘mellemlang videregående uddannelse’), the employment rate was 82% (females: 84%). For those with a long cycle higher education (MA or equivalent/higher), the employment rate was 79% (females: 77%). (p.65 unless otherwise specified)

*When you look at the descendants of non-Western immigrants at the age of 30 years, 41% of males and 25% of females have only a grundskoleuddannelse. The corresponding numbers for males and females of Danish origin are 18% and 13%. 22% of male- and 30% of female descendants of non-Western immigrants have a vocational education at the age of 30; the corresponding numbers for people of Danish origin are 40% and 30%. When it comes to medium-cycle higher education, the numbers for non-Western descendants are 6% and 15%; the corresponding numbers of people of Danish origin are 10% and 24%. 10% of male descendants and 8% of female descendants of non-Western immigrants at the age of 30 have a long cycle higher education; 13% of males of Danish origin and 15% of females of Danish origin at that age have one. As mentioned above there’s generally a pronounced gender difference when it comes to the education of non-Western descendants, as 61% of female descendants and 44% of male descendants at the age of 30 have a ELVQ. (p.67)

*I’ll add a couple of cautious remarks here regarding how to interpret the numbers above, cautious remarks which are not included in the report (so no page references): a) There’s probably a significant power issue here when considering forecasting based on these numbers, because the number of non-Western descendants in this age group (30-years-old) is quite low – n=558 (males) and n=559 (females). b) In terms of forecasting, heterogeneity might also be an issue. It matters if you’re looking at descendants born before or after 1983-84, because the composition of new immigrants changed at that point (in the medium run, so did the composition of immigrants in Denmark as a whole). I already talked a bit about related matters in the comment section here. Non-Westerns who came before, say, 1980 mostly came here to work; on the other hand the number of non-Westerns with fugitive status or family reunification status increased dramatically after 1983 due to policy changes implemented at that point. Another dimension along which heterogeneity is relevant is the change in the country profile of descendants, change which is not only driven by a change in the immigration patterns but also related to fertility differences across subpopulations; the total fertility rate of Somali immigrants is almost twice that of Turkish immigrants (86% higher, p.26) and these differences aren’t new. It should perhaps be made clear here that even if the change in the composition of non-Western descendants in the past might have had adverse effects on some human capital measures (SES of parents, IQ…) of the descendant group ‘as a whole’, it’s far from certain that this will lead to lower educational outcomes of the group in the future – for example, political commitment to improve educational outcomes of these groups might more than make up for the other effects. From 2004 to 2011 the educational outcomes of non-Western descendants improved, but there were only 72 non-Western descendants altogether in 2004 so it’s hard to draw strong conclusions from this as we once again run into the power issue.

*One way to try to draw inferences about the future educational profiles is to look at the educational profile of descendants currently aged 20-30 years old and compare them with the historical educational profiles of the 1980-generation (the current 30-year-olds). This is done below, the first graph contains data for the current 20-30 year-olds, the second contains data for the current 30-year-olds, green = females, blue = males – the lower ones are for non-Westerns, the graphs show how big a percentage of the group had obtained an ELVQ at any given age between 20 and 30. For example, 40% of non-Western males have an ELVQ at the age of 28 (and this was also the case for the 1980-generation):

*Part of the reason why I’ve focused mostly on descendants is that it is very hard to figure out the education levels of (first-generation) immigrants, because the data the authors made use of includes only educations which are completed at Danish educational institutions. In other words, both an Italian nuclear physicist educated in Rome and a poor Sudanese woman without a primary school education will have an ‘unknown’ education level (uoplyst) in these data sets, making it harder to pinpoint just exactly what is going on. A big majority of immigrants do not have a Danish education – 77% of Western and 69% of non-Western immigrants do not have a Danish education. (p.80) However, it seems relatively clear that at least when dealing with non-Western immigrants, an ‘unknown’ education level probably most often translates to a ‘low education level’ – the employment rate of non-Western female immigrants with an unknown education level is just 33% (p80).

January 25, 2012 Posted by | Data, Demographics, denmark, Economics, education, immigration | Leave a comment

Quotes

i. “Trying is the first step toward failure.” Homer Simpson.

ii. “Truth and clarity are complementary.” Niels Bohr.

iii. “If you can’t read and write you can’t think. Your thoughts are dispersed if you don’t know how to read and write. You’ve got to be able to look at your thoughts on paper and discover what a fool you were.” Ray Bradbury. Me: “were”?

iv. “one reason to move to evidence-based practice is that doctors aren’t trained as scientists. If anything, most medicine seems more of a craft than a science. That’s not to say that the practice of medicine doesn’t benefit from science; of course it does. But doctors aren’t trained to evaluate the science behind what they’re told; certainly not as well trained as those who actually do the research.

My father did research biology in a medical school. He tended to regard most doctors as plumbers; admittedly very useful ones to have around. But what they did wasn’t, to his mind, anything akin to scientific research or anything that might lead to the ability to evaluate same.” – Robert Levine. Related link. Also, this.

v. “It is well, when judging a friend, to remember that he is judging you with the same godlike and superior impartiality.” Arnold Bennett.

vi. “The traditional way of thinking about learning at a university is: there’s somebody who’s a teacher, who actually has some amount of knowledge, and their job is figuring out a way of communicating that knowledge. That’s literally a medieval model; it comes from the days when there weren’t a lot of printed books around, so someone read the book and explained it to everybody else. That’s our model for what university education, and for that matter high school education, ought to be like. It’s not a model that anybody’s ever found any independent evidence for.” Alison Gopnik, via John Hawks.

vii. “All opinions are not equal. Some are a very great deal more robust, sophisticated and well supported in logic and argument than others.” Douglas Adams.

viii. “If a little knowledge is dangerous, where is the man who has so much as to be out of danger?” T. H. Huxley. His advice? “Try to learn something about everything and everything about something.”

ix. “For a long time it has seemed to me that life was about to begin – Real Life. But there was always some obstacle in the way, something to be got through first, some unfinished business, time still to be served, a debt to be paid – then life would begin. At last it dawned on me that these obstacles were my life.” Alfred d’Souza.

x. “Power calls to those who are hungry for power, and there are hungry idiots everywhere.” Laura Anne Gilman.

xi. “Few men have virtue to withstand the highest bidder.” George Washington.

xii. “Though familiarity may not breed contempt, it takes off the edge of admiration.” William Hazlitt

xiii. “He will never have true friends who is afraid of making enemies.” -ll-

xiv. “Men of genius do not excel in any profession because they labour in it, but they labour in it because they excel.” -ll-

xv. “Great thoughts reduced to practice become great acts.” -ll-

xvi. “Modesty is the lowest of the virtues, and is a real confession of the deficiency it indicates. He who undervalues himself is justly undervalued by others.” -ll-

xvii. “Life is a sexually transmitted terminal disease.” Lewis Grizzard. The quote isn’t correct but if you add ‘human’ in front of life – which makes sense given the target group – I guess it works. It’s easy to forget but if you look at the entire history of life on Earth, that whole sexual reproduction-thing was really quite late to the party.

November 24, 2011 Posted by | education, Quotes/aphorisms | 3 Comments

Female education, some data

From the World Bank indicators. This is not how I’d have liked it to look like, but it’s exceedingly difficult to import data representations from the site to a blog like this, anyway I didn’t figure out how to do it and thought I’d spent too much time on it not to at least post this. Click to view in a much higher resolution:

“Primary completion rate is the percentage of students completing the last year of primary school. It is calculated by taking the total number of students in the last grade of primary school, minus the number of repeaters in that grade, divided by the total number of children of official graduation age.” The data is here.

I just took a look at some specific countries I was curious about for which data were available (there are lots of gaps in the data here, as you can tell from the graph). The world average was 75% in 1991, 78,9 in 2000 and 87,3 in 2009. Which kinda puts Afghanistan’s number of 21 % from 2005 into perspective. Note that the remainder here isn’t the number of females who don’t get a high school diploma (or equivalent); it is females who most likely haven’t really learned how to read. In case you were wondering, I did look at Somalia as another country example – this is one of those countries for which there are no data. Countries like these are not included when calculating the averages so the averages of the measures here are the high bars for these numbers, not the low bars.

The Indicators also have numbers on the gender ratios of variables like these, and that’s probably a better variable if you want to figure out if a particular country having a very low score is just an ‘ordinary’ s*#¤hole country where nobody can afford to go to school; or if it is a maledominated version of same where females just aren’t ever given the chance. Note that in some s*¤$hole countries, it’s probably the case that these numbers don’t look all that bad because the gender discrimination doesn’t take place at this specific level, but instead only kicks in later on (i.e. related to secondary or tertiary education). If you want to take a closer look at these cases there’s data for that stuff too, for example the ‘Ratio of female to male primary enrollment (%)’, ‘Progression to secondary school, female (%)’ and similar variables.

November 2, 2011 Posted by | Data, Demographics, education | Leave a comment

Some numbers

I spent a bit of time at Statistikbanken (Statbank Denmark) yesterday, below are some numbers from it that might be of interest. When you click the link you get to the front page of the site – now, if you look to the right there’s a small Union Jack which says ‘English’ if you hover over it. Click this and you get to the English version of the site. I don’t think all of the stuff at the Danish version of the site has been translated at the English link – but a lot of stuff has, so if you’re a foreigner curious about Denmark and the Danes, go take a look..

i. This part contains data from ‘KRHFU1: Befolkningens højeste fuldførte uddannelse (15-69 år) efter område, herkomst, uddannelse alder og køn’.

In 2010, when looking at the age segment of Danes who were 30-34 years old, 20494 Danish males and 22812 Danish females had as the highest achieved education level completed a ‘long-cycle higher education’ (I think this is the term they use in the English version of the data; in Danish it’s just ‘lang videregående uddannelse’. It corresponds to an education level above BA-level but below PhD-level, i.e. Master’s Degree or equivalent). Notice that more females than males at that age has completed this level of education. This is also true after you correct for the fact that there are more males than females in that age segment of the population; in total, there were 177078 males and 176291 females in that age segment of the Danish population. In terms of percentages of the total population in the specific age segment, 11,6 % of the males and 12,9 % of the females at the age of 30-34 had completed a long-cycle higher education in 2010 – the gender difference is about 10 percent.

Now, a funny thing happens when you compare these numbers to the age segment of Danes at the age of 65-69 (people who’ve just retired). In that sample, 9655 males and 3818 females have a long-cycle higher education – out of 146029
males and 152812 females. In that sample, 6,6 % of the males and just 2,5 % of the females have a long-cycle higher education – males in that age group are more than 2,5 times as likely to have a high education than females.

How does it look when you include the age groups in between those two? Like this:

More females than males get a long education today and it’s been that way for at least 10-15 years.

ii. This part contains data from ‘Folketal pr. 1. januar efter tid, alder og køn’ and ‘KM6: Befolkningen 1 januar efter kommune, køn, alder og folkekirkemedlemsskab’

(red: females, blue: males. The x-axis is age, the y-axis is the percentage of each age group who are members of Folkekirken)

So I took out the number of male and female members of Folkekirken at the ages of 1-80 and divided by the total number of Danes in the specific age-group – this gives a measure of how big a percentage of each age group is a member of Folkekirken (Danish National Church). It seems that there are some age cycles here. I did a quick logical test in Excel to get an overview of how the membership rate changes from age group to age group. At the ages of 1-15 years, membership grows ‘every year’ (2-year olds are more likely to be members than 1-year olds, ect.). At the age group of people 18-27 years old, membership drops ‘every year’. Between 30-43 it pretty much grows every year again, then it stabilizes around the new level. For people above the age of 55, it pretty much grows every year again. I decided to not include people above the age of 80 because nothing much of interest happens there; as should be clear from the graph this age segment has by far the highest membership rates and more than 9 out of 10 are members. Remember when interpreting the relatively low membership of children to the left of the graph and the membership growth of the 1-15 years old that part of this is probably because of the relatively higher fertility of Muslim immigrants (as opposed to fewer atheist children).

iii. This part contains data from ‘FAM55N: Husstande pr. 1. januar efter kommune/region, husstandstype og husstandsstørrelse’. Every time some econ blog posts something about the household income development over time (like this one) I also see a commenter asking: ‘but what about household size?’ What I very rarely see is a commenter linking to actual data on household size. This puzzles me every time, because at least in Denmark that kind of data actually isn’t all that hard to get your hands on. Here’s a quick run from Statistikbanken:

I omitted some of the classes because otherwise it quickly gets very messy and they don’t add much to the big picture anyway, this is why the numbers don’t quite add up to the total population – but the table does include far most Danes (the 2011 numbers include 4,92 million people, the 1986 numbers 4,42 million people). The number of single person households with one male or one female living alone has increased somewhat. If you wanted to do it completely right, you’d add all the omitted classes as well before making the calculation, but in terms of the people in the sample (which covers ~ 90% of all Danes) the percentage of people living in single person households went up from 16,2 % to 20,3 %. In terms of the percentage of all households that are single person households, the number is of course much higher. In 1986, 35,6 % of all households (in the sample) were single person households, in 2011 it was 41,5 %. The number has gone up, but less than I’d thought.

I found it interesting that the number of households with a married couple and 3-4 inhabitants altogether (the most likely constellation is a married couple plus 1 or 2 children) has decreased significantly and movement from ‘married couples’ to ‘other couples’ does not explain all of it. Is the driver an increase in the divorce rate or lower fertility rate? I don’t know.

September 14, 2011 Posted by | Data, Demographics, Economics, education | Leave a comment

Some basic stuff (mostly for the non-economists)

As a student of “economics”, I’m pretty sure I’ve had more courses dealing with stuff like this (and of course stuff a bit harder) than I’ve had courses about how ‘the economy’ supposedly ‘works’.

In related news, semester started yesterday so I might update a bit less frequently in the time to come. We’ll see how it goes.

August 31, 2011 Posted by | education, Khan Academy, knowledge sharing, Mathematics, Statistics | Leave a comment

Cosmology revisited

So first of all, I know a handful readers or two came by after I commented over at William’s blog – if one or more of you decided to come back to read this: Welcome!

If you didn’t read this post (that is: looked closely at the images) back when I posted that, I suggest you start there. Now Salman Khan has made a series of videos where he starts at Earth, then moves on outwards. I notice in one of the videos he mistakenly uses light year as a measure of time, not distance, but he was pretty excited at that point, for good reason. I’ve posted the first video in the series below – when I watched it on youtube, it automatically started the next video once the previous one had finished, which was both good and bad as I probably sat there for over an hour watching that stuff, but I don’t know if it’ll do the same when embedded here. If not, you should really watch the series on youtube if you think the first part was ok – it gets even better and far more mind boggling as he proceeds.

I love what Sal is doing. If you felt the need to follow the link to Salman Khan’s wikipedia article because you don’t know who he is or what he’s doing, here’s another good video you should watch:

And here’s the link to the site.

In other news, here’s a chess game I played earlier this evening (I was white and it was a 5 minute game so presumably lots of mistakes if you let the silicon monster have a look at it). I haven’t run it through a computer, but I still think my decision to exchange on g7 and move 20.f5 instead of taking on e6 was the right one. I really liked that 20.f5 move when I played it. If black wants to survive, he can’t defend that e6 pawn anyway, i.e. 20…Nf8, 21.f6+ Kg8, 22.Qd2 Nbd7, 23.Qh6 Nf6 (…Ne6, 24.dxe6 Nxf6(□), 25.Nxf6+ Qxf6(□), 26.Rxf6 and white has the same win as in the game with the Nf3 and Ng5-manoeuvre), 24.Nxf6+ Kg8, 25.Nh5! Ne6 (…gxh5 and after 26.Rxf7 black is mated), 26.dxe6 Rg8(□), 27.exf7 and game over). I think 20.f6 was a better defence than Ne5, Ne5 was a bad move. Black needs all the support he can get of the black squares around his king after he’s allowed the exchange of the g7-bishop. That said, the position after f6 is still losing for black.

June 12, 2011 Posted by | Astronomy, Chess, cosmology, education | Leave a comment

The US primary education system

Of course it doesn’t tell the whole story, but the story Steven Brill tells in this article is so insane anyway that you probably don’t need to be reminded of that.

No excerpt does justice to the piece, read all of it – and yes, I know I’m late to the party, it’s an old article and perhaps some of you already have. Megan McArdle has more/new related stuff here, there’s also quite a bit of comments to read through if you have the time – apparently the ‘rubber rooms’ are no more, but as to the other stuff I don’t know how much has happened… A few bits from Brill’s article:

“Brandi Scheiner says that her case is likely to be heard next year. By then, she will have twenty-four years’ seniority, which entitles her to a pension of nearly half her salary—that is, her salary at the time of retirement—for life, even if she is found incompetent and dismissed. Because two per cent of her salary is added to her pension for each year of seniority, a three-year stay in the Rubber Room will cost not only three hundred thousand dollars in salary [she currently earns more than a hundred thousand dollars a year] but at least six thousand dollars a year in additional lifetime pension benefits.” […]

“Scheiner refused to allow me access to the complete file related to her incompetence proceeding, which would detail the charges against her and any responses she might have filed, saying only that “they charged me with incompetence—boilerplate stuff.” (Nor could Felder comment, because Scheiner had insisted that her file be kept sealed.) But Scheiner did say that she and several of her colleagues in the Rubber Room had brought a “really interesting” class-action suit against the city for violations of their due-process and First Amendment rights as whistle-blowers. She said that the suit was pending, and that she would be vindicated. Actually, she filed three suits, two of which had long since been dismissed. And, a month and a day before she mentioned it to me, the magistrate handling the third case—in a move typically reserved for the most frivolous litigation—had ordered Scheiner and her co-plaintiffs to pay ten thousand dollars to the city in court costs, because that filing was so much like the second case. This third case is pending, though it no longer has a lawyer, because the one who brought these cases has since been disbarred, for allegedly lying to a court and allegedly stealing from Holocaust-survivor clients in unrelated cases.

It takes between two and five years for cases to be heard by an arbitrator”

“Mohammed’s case was the first to reach arbitration since the introduction of an initiative called Peer Intervention Program (P.I.P.) Plus, which was created to address the problem of tenured teachers who are suspected of incompetence, not those accused of a crime or other misconduct. […] The deal seemed good for both sides: a teacher accused of incompetence would first be assigned a “peer”—a retired teacher or principal—from a neutral consulting company agreed upon by the union and the city. The peer would observe the teacher for up to a year and provide counselling. If the observer determined that the teacher was indeed incompetent and was unlikely to improve, the observer would write a detailed report saying so. The report could then be used as evidence in a removal hearing conducted by an arbitrator agreed upon by the union and the city. […] Under the union contract, hearings on each case are held five days a month during the school year and two days a month during the summer. Mohammed’s case is likely to take between forty and forty-five hearing days—eight times as long as the average criminal trial in the United States. […] When the bill for the arbitrator is added to the cost of the city’s lawyers and court reporters and the time spent in court by the principal and the assistant principal, Mohammed’s case will probably have cost the city and the state (which pays the arbitrator) about four hundred thousand dollars.

Nor is it by any means certain that, as a result of that investment, New York taxpayers will have to stop paying Mohammed’s salary, eighty-five thousand dollars a year. Arbitrators have so far proved reluctant to dismiss teachers for incompetence. […] in the past two years arbitrators have terminated only two teachers for incompetence alone, and only six others in cases where, according to the Department of Education, the main charge was incompetence.”

“[a study] examined teacher rating processes, and found that in districts that have a binary, satisfactory-unsatisfactory system, ninety-nine per cent of teachers receive a satisfactory rating, and that even in the few school districts that attempt a broader range of rating options ninety-four per cent get one of the top two ratings.”

March 8, 2011 Posted by | education | 2 Comments

Random thoughts on education

A society where people finance higher education at least in part by personal debt will on average reward individuals who are risk-takers and punish individuals who are risk-averse, relative to a society where (some higher) education is fully financed by the government.

I know that it might be offset at least in part by future wage increases, but the reform suggestions that are rumoured (Danish link) will rob me of maybe 150k kroners over the next few years. No, it was never my money to begin with, but… My brother bought a car when he got a job after his studies, the value of that car corresponds roughly to the value of that money*.

In principle maybe such a reform is a good idea, but I can think of a whole lot of other ways for the government to try to solve the main problem, which is that thousands of young people spend years taking educations that teach them nothing useful and have a negative ROI. Of course that problem will not become fully apparent until the government tries to actually solve it, but anyway, my best guess is that some people will drop out, the unemployment numbers will go up and the government will have to think up a new scheme to better hide that unemployment five-ten years down the line (a scheme like efterløn – anybody remember that one? Hopefully not, that’s part of why the government is discussing this reform. 150k probably(?) also roughly corresponds to one year’s worth of government payments of efterløn to a 63-year old. But there are a lot of those people and they all vote – and anyway most students already hate this government so there’s no need to try to please them).

An MA will be worth more after the reform relative to now, so implicitly this transition will correspond to a transfer from future students to people who currently already have a high education (and wage) – the latter get lower wage- and employment competition from people with the same education level in the long run, which’ll result in a higher wage.

The current (proposed) model says that people are only to pay back the loan once they are considered high earners – that is, once they hit the top marginal tax rate – is that an incentive to stay below the income level of the topskat-limit or what? It will cause some people close to the limit to work less than they otherwise would, and much less than they would have if there was both no such high-earner clause and no ‘topskat’, and it’ll of course make the size of the education premium even more difficult to figure out than it is now. Yes, I know, of course bracket creep will solve that problem in the long run. Most if not all these things will come as a complete surprise to the people who vote for this law, or so they’ll claim.

When I initially learned about the reform proposals, my first thought was something along the lines of “that’s too bad, now I’ll have to drop out and I’ll never get my MA, even if I’ve worked quite hard getting back on my feet after a couple of hard years”. Now I’m at least considering trying to finish my education even if that stuff goes through, primarily because of the ‘only high earners pay back the loan’-clause (which is not all bad, though it is also quite bad, for the reasons mentioned above) that I’d overlooked. Remove that one and I’d be quite likely to leave the university next semester if it goes through.

*to foreigners like Plamus: No, he didn’t buy a brand new BMW – and he lent the money to pay for the car, he of course didn’t have that kind of money at the time. Cars are expensive in Denmark. Not pre-tax, pre-tax Denmark was the cheapest country in the European Union in 2007 (link). Post-tax it’s quite bad though. Here’s a page on the relevant taxation levels, by the ministry of taxation (Skatteministeriet). The example on that page features a car with a supplier pre-tax sales price of 85.000 kroners (appr. $16.000 using the current official exchange rate of 5,29 kroners/dollar) which ends up costing the consumer 222.850 kroners ($42k). No, don’t be deceived, it’s not that once you have the car, it’s cheap to drive around in it. A gallon of gasoline is currently at about 7,5 dollars (10,5 kroners/liter). Comparing that to these prices, well…

October 17, 2010 Posted by | education, Personal | 2 Comments

Superfreakonomics (1,5?)

Have read the first 100 pages now, a few interesting passages from the book:

1) “In the early 1910s, the Department of Justice conducted a census of 310 cities in 26 states to tally the number of prostitutes in the United States: “We arrive at the conservative figure of approximately 200,000 women in the regular army of vice.”
At the time, the American population included 22 million women between the ages of fifteen and forty-four. If the DOJ numbers are to be believed, 1 og every 110 women in that age range was a prostitute. But most prostitutes, about 85 percent, were in their twenties. In that age range, 1 of every 50 American women was a prostitute.
The market was particularly strong in Chicago, which had more than 1000 known brothels.”

2) “Teaching has traditionally been dominated by women. A hundred years ago, it was one of the few jobs available to women that didn’t involve cooking, cleaning, or other menial labor. (Nursing was another such profession, but teaching was far more prominent, with six teachers for every nurse.) At the time, nearly 6 percent of the female work-force were teachers, trailing only laborers (19 percent), servants (16 percent), and laundresses (6,5 percent). […] As of 1940, an astonishing 55 percent of all college-educated female workers in their thirties were employed as teachers.”

Yes, the teacher numbers are interesting, but damn, so are the others – a 100 years ago, appr. 1 in 6 women employed were servants! Here’s a bit more:

“In 1960, about 40 percent of female teachers scored in the top quintile of IQ and other aptitude tests, with only 8 percent in the buttom. Twenty years later, fewer than half as many were in the top quintile, with more than twice as many in the buttom.”

Combine this fact and the development since then with the Flynn effect and you get at least part of the explanation why teachers were more respected in the past. They, and especially the females, were much smarter on average.

3) “Alan Krueger combed through a Hezbollah newsletter called Al-Ahd (The Oath) and compiled biographical details on 129 dead shahids (martyrs). He then compared them with men from the same age bracket in the general populace of Lebanon. The terrorists, he found, were less likely to come from a poor family (28 percent versus 33 percent) and more likely to have at least a high-school education (47 percent versus 38 percent).
A similar analysis of Palestinian suicide bombers by Claude Berrebi found that only 16 percent came from impoverished families, versus more than 30 percent of male Palestinians overall. More than 60 percent of the bombers, meanwhile, had gone beyond high school, versus 15 percent of the populace.
In general, Krueger found, “terrorists tend to be drawn from well-educated, middle-class or high-income families”.”

4) “The beauty of terrorism – if you’re a terrorist – is that you can succeed even by failing. […] Let’s say it takes an average of one minute to reomve and replace your shoes in the airport security line. In the United States alone, this procedure happens roughly 560 million times per year. Five hundred and sixty million minutes equals more than 1,065 years – which, divided by 77.8 years (the average U.S. life expectancy at birth), yields a total of nearly 14 person-lives. So even though Richard Reid [the failed shoe bomber] failed to kill a single person, he levied a tax that is the time equivalent of 14 lives per year.”

5) “Until the 1960s, hospitals simply weren’t designed to treat emergencies. “If you brought someone to a hospital at night,” Feied [Craig Feied, according to the book ‘an emergency-medicine specialist’] says, “the doors would be locked. You’d ring the bell, a nurse would come down to see what you wanted. She might let you in, then she’d call the doctor at home, and he might or might not come in.” Ambulances were often run by the local mortuary. It is hard to think of a better example of misaligned incentives: a funeral director who is put in charge of helping a patient not die!”

[…]

“In a given year, an excellent ER doctor’s patients will have a twelve-month death rate that is nearly 10 percent lower than the average. This may not sound like much, but in a busy ER with tens of thousands of patients, an excellent doctor might save six or seven lives a year relative to the worst doctor.”

‘No, it doesn’t sound like much and that really isn’t a big difference’ I’d say, unless you’re one of those six or seven marginal people of course. But the difference between the best and an average ER doctor seems to be quite small – it’s probably much bigger in other specialties. A Dane at the age of 30 can expect to live 8,5 years more if he’s never smoked than if he’s a heavy smoker, so a GP who’s good at making people stop smoking will save dozens of lives every year on that count alone.

6) “More than $40 billion is spent worldwide each year on cancer drugs.”

(I’d have thought the number was bigger than that)

“cancer patients make up 20 percent of Medicare cases but consume 40 percent of the Medicare drug budget.”

“A typical chemotherapy regime for non-small-cell lung cancer costs more than $40,000 but helps extend a patient’s life by an average of just two months.”

I like the book so far.

August 16, 2010 Posted by | Books, Cancer/oncology, Economics, education, History, IQ, Medicine | Leave a comment

College degrees and gender

Since 1970, the gender ratio has flipped:

collegedegrees

Only business, engineering and the physical sciences have more males than females enrolled. Via Mark Perry.

October 27, 2009 Posted by | Data, Demographics, education, USA | Leave a comment

If research papers had a comment section…

phd052709s

Link.

June 15, 2009 Posted by | education, fun | Leave a comment