Large US-representative adolescent sample: A Flynn Effect was found for IQs ≥ 130, a negative one for ≤ 70; this challenge the practice of generalizing IQ trends with non-representative data samples

The Flynn effect for fluid IQ may not generalize to all ages or ability levels: A population-based study of 10,000 US adolescents. Jonathan M. Platt et al. Intelligence, Volume 77, November–December 2019, 101385. https://doi.org/10.1016/j.intell.2019.101385

Highlights
• When outdated norms are used, the Flynn Effect inflates IQs and potentially biases intellectual disability diagnosis
• In a large US-representative adolescent sample, a Flynn Effect was found for IQs ≥ 130, and a negative effect for IQs ≤ 70
• IQ changes also differed substantially by age group
• A negative Flynn Effect for those with low intellectual ability suggests widening disparities in cognitive ability
• Findings challenge the practice of generalizing IQ trends based on data from non-representative samples

Abstract: Generational changes in IQ (the Flynn Effect) have been extensively researched and debated. Within the US, gains of 3 points per decade have been accepted as consistent across age and ability level, suggesting that tests with outdated norms yield spuriously high IQs. However, findings are generally based on small samples, have not been validated across ability levels, and conflict with reverse effects recently identified in Scandinavia and other countries. Using a well-validated measure of fluid intelligence, we investigated the Flynn Effect by comparing scores normed in 1989 and 2003, among a representative sample of American adolescents ages 13–18 (n = 10,073). Additionally, we examined Flynn Effect variation by age, sex, ability level, parental age, and SES. Adjusted mean IQ differences per decade were calculated using generalized linear models. Overall the Flynn Effect was not significant; however, effects varied substantially by age and ability level. IQs increased 2.3 points at age 13 (95% CI = 2.0, 2.7), but decreased 1.6 points at age 18 (95% CI = −2.1, −1.2). IQs decreased 4.9 points for those with IQ ≤ 70 (95% CI = −4.9, −4.8), but increased 3.5 points among those with IQ ≥ 130 (95% CI = 3.4, 3.6). The Flynn Effect was not meaningfully related to other background variables. Using the largest sample of US adolescent IQs to date, we demonstrate significant heterogeneity in fluid IQ changes over time. Reverse Flynn Effects at age 18 are consistent with previous data, and those with lower ability levels are exhibiting worsening IQ over time. Findings by age and ability level challenge generalizing IQ trends throughout the general population.

Keywords: IntelligenceFlynn effectAdolescenceIntellectual disabilities

Cool charts and tables at the publisher's link above. Excerpts:

5. Discussion

The present study utilized data from a large US-representative
sample of adolescents to describe changes in IQ between 1989 and
2003. There were three central findings: 1) Overall, there was no evidence
of a Flynn Effect during the study period; 2) however, overall IQ
trends masked substantial heterogeneity in the presence and direction
of the Flynn Effect by both ability level and age; and 3) there was no
variation in the Flynn effect as a function of other sociodemographic
characteristics.
The overall lack of a Flynn Effect in our sample is concordant with
trends in the K-BIT, KBIT-2, the Kaufman Assessment Battery for
Children (K-ABC and KABC-II), and other individually administered
screening tests reported in a previous meta-analysis (Trahan et al.,
2014). It also conforms with the conclusion that gains have decreased
in more recent decades (Pietschnig & Voracek, 2015). However, studies
using other tests (e.g., Wechsler scales) did find substantial Flynn
Effects (Pietschnig & Voracek, 2015; Trahan et al., 2014). Explanations
for the Flynn Effect are diverse. Although genetic explanations focusing
on factors such as hybrid vigor (Mingroni, 2007; Rodgers & Wänström,
2007) have been proposed, environmental explanations predominate
(Dickens & Flynn, 2001), emphasizing societal changes in perinatal
nutrition (Lynn, 2009) and nutrition in general (Colom, Lluis-Font, &
Andrés-Pueyo, 2005), education (Teasdale & Owen, 2005), reduced
number of siblings (Sundet, Borren, & Tambs, 2008), the prevalence of
parasites and the burden of disease (Daniele & Ostuni, 2013; Eppig,
Fincher, & Thornhill, 2010), and increased environmental complexity
(Schooler, 1998).
By contrast, other studies have reported reverse Flynn Effects. In
discussing these negative trends in Scandinavian countries, Lynn and
colleagues hypothesized that they may be due to greater fertility among
low SES groups, immigrants, and older adults (Dutton et al., 2016;
Dutton & Lynn, 2013). However, a recent analysis in Norway to test
these claims largely rejects their hypotheses, reporting that Flynn
Effects were not consistent within families over time (Bratsberg &
Rogeberg, 2018). Further, a recent meta-analysis found no substantial
role of fertility on test score changes across an array of studies
(Pietschnig & Voracek, 2015), and recent empirical evidence suggests
that immigration effects do not play a meaningful role in explaining
Flynn Effect reversals (Pietschnig, Voracek, & Gittler, 2018).
We add to the evidence reported in previous studies, by reporting
heterogeneity in the Flynn Effect by ability level and age. We find
support for a reverse Flynn Effect for those of low ability and older age,
and a positive Flynn Effect for those of high ability and younger age.
These results have several implications. First, they signal a widening
disparity in the US in terms of cognitive ability, with those at the lower
end of the ability dimension not only exhibiting less gains than those at
the higher ends, but reversing direction entirely. Second, these results
have implications for considering demographic differences when adjusting
IQ test scores in the population.
Improvements in education, nutrition, prenatal and post-natal care,
and overall environmental complexity over the past century are
thought to contribute to the Flynn Effect in the overall population
(Dickens & Flynn, 2001; Lynn, 2009; Schooler, 1998; Teasdale & Owen,
2005). However, the disparities by ability level that we identified
suggest that the benefits from these societal improvements have been
more dramatic for those at the highest ability levels, potentially because
they are better able to take advantage of these societal changes. This
interpretation is in line with Fundamental Cause Theory (Phelan, Link,
& Tehranifar, 2010), which argues that when new knowledge or technology
is introduced into a society, those with the highest status are
most likely to take advantage first and benefit. Disproportionate utilization
by those with higher abilities may widen intellectual disparities,
leaving those at the lowest ability levels worse off than before. We note,
however, that the Flynn Effect did not differ across other measures of
status, such as poverty and parental education. The correlation analyses
we conducted revealed a positive association of moderate magnitude
between IQ and the size of Flynn Effect, for every age group between 13
and 18, regardless of whether that group showed an overall positive or
negative Flynn Effect. One possible interpretation of this pattern is that
adolescents with high fluid intelligence, not necessarily those with the
highest access to resources, have benefitted most from societal progress
over time.
Previous research on the stability of the Flynn Effect across ability
levels has produced inconsistent and inconclusive results (McGrew,
2015; Weiss, 2010). Sometimes it has been higher at low IQs, and
sometimes a reverse Flynn Effect has been found in high IQ samples
(Spitz, 1989; Teasdale & Owen, 1989; Zhou et al., 2010). A meta-analysis examining ability level as a moderator variable did not observe a
Flynn Effect for those with low IQ (Trahan et al., 2014). However,
previous studies differ in quality (Trahan et al., 2014) and often rely on
small sample sizes at the lower end of the IQ distribution (Zhou et al.,
2010). Specifically, Trahan and colleagues noted, “the distribution of
Flynn effects that we observed at lower ability levels might be the result
of artifacts found in studies of groups within this range of ability” (p.
1349).
We also identified variation in the Flynn Effect by age. The positive
Flynn Effect of 2.3 points per decade at age 13 approximately equals the
value obtained in a summary of studies of Raven's matrices for nearly
250,000 children in 45 countries (Brouwers, Van de Vijver, & Van
Hemert, 2009) and in a meta-analysis of about 14,000 children and
adults in the US and UK (Trahan et al., 2014). However, the 2-point
value is smaller than the traditional 3 points for global intelligence and
4 points for fluid intelligence (Pietschnig & Voracek, 2015). Likewise,
the reverse Flynn Effect that occurred at ages 15–18 was similar to
effects reported in Scandinavian countries among young adult males
during the same time period (Bratsberg & Rogeberg, 2018; Dutton &
Lynn, 2013; Sundet et al., 2004; Teasdale & Owen, 2005, 2008), and in
other countries as well, such as France (adults tested on WAIS-III and
WAIS-IV) and Estonia (young adults tested on Raven's Matrices)
(Dutton et al., 2016). The age effects are discordant with previous
metaanalyses. Pietschnig and Voracek (2015) evaluated age effects and
found stronger gains for adults than children. In their meta-analysis,
Trahan et al. (2014) did not find a significant relationship between
Flynn Effect and age in their examination of the mean ages across
heterogeneous and often small samples. Our methodology differed from
the techniques used in both meta-analyses, as we studied large samples
that were homogeneous by age.
The notable differences we identify among narrowly defined age
groups may be related to cognitive and neurodevelopmental changes
that occur during adolescence. Fluid reasoning abilities and cognitive
abilities that support reasoning (e.g., rule representation) develop rapidly during early adolescence (Crone et al., 2009; Crone, Donohue,
Honomichl, Wendelken, & Bunge, 2006; Ferrer, O'Hare, & Bunge, 2009;
Žebec, Demetriou, & Kotrla-Topić, 2015). Brain regions that play a
central role in reasoning and problem solving, including the dorsolateral and ventrolateral prefrontal cortex and superior and inferior
parietal cortex, also exhibit dramatic changes in structure and function
across adolescence (Bunge, Wendelken, Badre, & Wagner, 2004; Ferrer
et al., 2009; Gogtay et al., 2004; Wendelken, Ferrer, Whitaker, & Bunge,
2015; Wright, Matlen, Baym, Ferrer, & Bunge, 2008). The notably different Flynn Effects by age in our study caution against generalizing
findings for a specific sub-group (such as conscripted young adult
males, which comprise the Scandinavian samples) to the nation as a
whole (Dutton & Lynn, 2013).
The present study identified no meaningful relationship between
Flynn Effect and poverty, parental education other sociodemographic
variables and background factors, including parental nationality, birth
order, family size, age of birth mother and father. This finding is notable given that these demographic variables are associated with IQ
level (von Stumm & Plomin, 2015), including in our sample (Platt,
Keyes, et al., 2018).
The results of this study should be considered in light of several
limitations. First, the study data were obtained 15 years ago. However,
this period was an ideal time to evaluate the presence of a reverse Flynn
Effect in the US, given the reverse effects found in Denmark, Norway,
Finland, and several other countries (Dutton et al., 2016; Teasdale &
Owen, 2008). In more recent years, no reverse Flynn Effect has been
observed for Wechsler's scales, as gains on the WAIS-IV (Wechsler,
2008) and WISC-V (Wechsler, 2014). Full Scale IQ have been close to
the hypothesized value of 3 points per decade (J Grégoire & Weiss,
2019; Jacques Grégoire, Daniel, Llorente, & Weiss, 2016 Weiss,
Gregoire, & Zhu, 2016; Zhou et al., 2010), especially when test content
is held constant (J Grégoire & Weiss, 2019; Weiss et al., 2016).
Second, the K-BIT nonverbal test is a screening test that measures a
single cognitive ability. It is, however, an analog of Raven's popular
matrices test which is commonly used in Flynn Effect studies (Brouwers
et al., 2009; Flynn, 1998; Pietschnig & Voracek, 2015). The Flynn Effect
is known to differ for different cognitive abilities (e.g., fluid intelligence, short-term memory) (Pietschnig & Voracek, 2015; Teasdale
& Owen, 2008), which may contribute to heterogeneity in findings
across studies with differing IQ measures. However, the K-BIT and
KBIT-2 nonverbal IQ is substantially correlated with comprehensive IQ
tests, such as the Wechsler's Full Scale IQ (mid-.50s to mid-70s)
(Canivez et al., 2005; Kaufman & Kaufman, 1990, 2004), though it is
lower than the correlation between different comprehensive test batteries (Kaufman, 2009; Wechsler, 2014). The present findings are descriptive and any practical application regarding the adjustment of IQs
must be made with the awareness that clinical diagnosis, such as the
identification of individuals with intellectual disabilities, must be based
on comprehensive IQ tests such as Wechsler's scales or the WoodcockJohnson, which assess multiple cognitive abilities.
Third, the study included only adolescents, which represents a
narrow period that may not capture meaningful developmental
changes. Indeed, fluid reasoning changes between ages 13–18 are
minimal (Wechsler, 2008, 2014), including in the present 2003 K-BIT
norms sample (Keyes et al., 2016) and the original 1989 norms sample
Kaufman & Kaufman (1990, Table 4.7). This age pattern may partially
explain why we found no overall Flynn Effect in this sample.
Fourth, different procedures were used to develop the 1989 and
2003 norms. The 1989 norms were estimated based on aggregated data
across all age groups, in order to stabilize norms at all ages (Angoff &
Robertson, 1987). Although slightly different statistical techniques
were used to develop the 2003 norms, the general approach to norms
development was similar between samples, and one test author (ASK)
was involved in the development of both sets of norms. Both samples
were representative of the US distributions of sociodemographic, economic, and other key background variables at the time (Kaufman &
Kaufman, 1990; Kessler, Avenevoli, Costello, et al., 2009). Further, both
sets of norms are based on six-month age bands. These samples are at
least as convergent as similar studies comparing samples used to develop original vs. revised norms. Previous studies have differed substantially by key sociodemographic distributions, such as the WISC and
WISC-R (Wechsler, 1949, 1974), which were key samples in the development of the Flynn Effect theory (Flynn, 1984). In the present
study, we adjusted the Flynn Effect for an array of background variables
to further minimize any differences between the 1989 and 2003 norms
samples that may confound the Flynn Effect estimates.
Fifth, the Flynn Effect has had a non-linear trajectory over the past
century (Pietschnig & Voracek, 2015). Because our study included IQ
measurements at only two time points, we were not able to test the
linearity of change over time.
This study is strengthened by the use of a large and representative
adolescent sample, with IQs measured with reasoning items that are
widely accepted as prototypical measures of fluid intelligence (Dutton
et al., 2016). The use of two sets of norms based on a single
administration of a test avoids practice effects and bias that may arise
from use of different versions of a test.
In conclusion, this study reports important heterogeneity in the
Flynn Effect among a nationally-representative sample of US adolescents.
We confirmed previous reports of reverse Flynn Effects among
large samples of older adolescent males, and extended the same pattern
to females. We also found important differential Flynn Effects by ability
level. These results add to a growing body of evidence suggesting that
Flynn Effect findings from narrow age bands or ability levels may
produce divergent findings that do not generalize to the overall population.
However, given the potential life or death implications of this
research in determining intellectual status in capital punishment cases,
the strength of evidence needed for definitive conclusions is extremely
high. At this time, we do not have sufficient evidence to recommend
differential adjustments to IQ scores. Additional research is needed to
replicate the current findings on the full age range and across comprehensive
measures of intelligence.

On psychological researcher's strategic behavior: High prevalence of effect declines with each new study of some question, yielding a ratio of 2:1; these declines are systematic, strong, and ubiquitous

Effect Declines are Systematic, Strong, and Ubiquitous: A Meta-Meta- Analysis of the Decline. Jakob Pietschnig et al. Front. Psychol., Nov 2019, doi: 10.3389/fpsyg.2019.02874

Abstract: Empirical sciences in general and psychological science in particular are plagued by replicability problems and biased published effect sizes. Although dissemination bias-related phenomena such as publication bias, time-lag bias, or visibility bias are well-known and have been intensively studied, another variant of effect distorting mechanisms, so-called decline effects, have not. Conceptually, decline effects are rooted in low initial (exploratory) study power due to strategic researcher behavior and can be expected to yield overproportional effect declines. Although decline effects have been documented in individual meta-analytic investigations, systematic evidence for decline effects in the psychological literature remains to date unavailable. Therefore, we present in this meta-meta-analysis a systematic investigation of the decline effect in intelligence research. In all, data from 22 meta-analyses comprising 36 meta-analytical and 1,391 primary effect sizes (N = 697,000+) that have been published in the journal Intelligence were included in our analyses. Two different analytic approaches showed consistent evidence for a higher prevalence of cross-temporal effect declines compared to effect increases, yielding a ratio of about 2:1. Moreover, effect declines were considerably stronger when referenced to the initial primary study within a meta-analysis, yielding about twice the magnitude of effect increases. Effect misestimations were more substantial when initial studies had smaller sample sizes and reported larger effects, thus indicating suboptimal initial study power as the main driver of effect misestimations in initial studies. Post-hoc study power comparisons of initial versus subsequent studies were consistent with this interpretation, showing substantially lower initial study power of declining, than of increasing effects. Our findings add another facet to the ever accumulating evidence about non-trivial effect misestimations in the scientific literature. We therefore stress the necessity for more rigorous protocols when it comes to designing and conducting primary research as well as reporting findings in exploratory and replication studies. Increasing transparency in scientific processes such as data sharing, (exploratory) study preregistration, but also self- (or independent) replication preceding the publication of exploratory findings may be suitable approaches to strengthen the credibility of empirical research in general and psychological science in particular.

Keywords: decline effect, meta-meta-analysis, Dissemination bias, effect misestimation, Intelligence

Not only assholes drive Mercedes; besides disagreeable men, also conscientious people drive high‐status cars

Not only assholes drive Mercedes. Besides disagreeable men, also conscientious people drive high‐status cars. Jan Erik Lönnqvist, Ville‐Juhani Ilmarinen, Sointu Leikas. International Journal of Psychology, December 3 2019. https://doi.org/10.1002/ijop.12642

Abstract: In a representative sample of Finnish car owners (N = 1892) we connected the Five‐Factor Model personality dimensions to driving a high‐status car. Regardless of whether income was included in the logistic model, disagreeable men and conscientious people in general were particularly likely to drive high‐status cars. The results regarding agreeableness are consistent with prior work that has argued for the role of narcissism in status consumption. Regarding conscientiousness, the results can be interpreted from the perspective of self‐congruity theory, according to which consumers purchase brands that best reflect their actual or ideal personalities. An important implication is that the association between driving a high‐status car and unethical driving behaviour may not, as is commonly argued, be due to the corruptive effects of wealth. Rather, certain personality traits, such as low agreeableness, may be associated with both unethical driving behaviour and with driving a high‐status car.

The introduction of a new machine translation system has significantly increased international trade on this platform, increasing exports by 10.9%; found too a substantial reduction in translation costs

Does Machine Translation Affect International Trade? Evidence from a Large Digital Platform. Erik Brynjolfsson, Xiang Hui, Meng Liu. Management ScienceVol. 65, No. 12, Sep 3 2019. https://doi.org/10.1287/mnsc.2019.3388

Abstract: Artificial intelligence (AI) is surpassing human performance in a growing number of domains. However, there is limited evidence of its economic effects. Using data from a digital platform, we study a key application of AI: machine translation. We find that the introduction of a new machine translation system has significantly increased international trade on this platform, increasing exports by 10.9%. Furthermore, heterogeneous treatment effects are consistent with a substantial reduction in translation costs. Our results provide causal evidence that language barriers significantly hinder trade and that AI has already begun to improve economic efficiency in at least one domain.


1.1. Related Literature and Contribution

1.1.1. AI and Economic Welfare. The current generation
of AI represents a revolution of prediction and classification
capabilities (e.g.,Brynjolfsson andMcAfee 2017).
Recent breakthroughs in ML, especially supervised
learning systems using deep neural networks, have
allowed substantial improvements in many technical
capabilities. Machines have surpassed humans at
tasks as diverse as playing the game Go (Silver et al.
2016) and recognizing cancer from medical images
(Esteva et al. 2017). There is active work converting
these breakthroughs into practical applications, such
as self-driving cars, substitutes for human-powered
call centers, and new roles for radiologists and pathologists,
but the complementary innovations required
are often costly (Brynjolfsson et al. 2019).
Machine translation has also experienced significant
improvement because of advances in ML. For
instance, the best score at the Workshop on Machine
Translation for translating English into German improved
from 23.5 in 2011 to 49.9 in 2018,2 according to
the widely used BLEU score, which measures how
close the MT translation output is to one or more
reference translations by linguistic experts (for details,
see Papineni et al. 2002). Much of the recent
progress in MT has been a shift from symbolic approaches
toward statistical and deep neural network
approaches. For our study, an important characteristic
of eMT is that replacing human translators with
MT or upgrading MT is relatively seamless. For instance,
for product listings on eBay, users consume
the output of the translation system but, otherwise,
need not change their buying or selling process. Although
users care about the quality of translation, it
makes no difference whether it was produced by a
human or machine. Thus, adoption of MT can be very
fast and its economic effects, especially on digital
platforms, immediate. Although, so far, much of the
work on the economic effects of AI has been theoretical
(Sachs and Kotlikoff 2012, Aghion et al. 2017,
Korinek and Stiglitz 2017, Acemoglu and Restrepo
2018, Agrawal et al. 2019) and notably (Goldfarb and
Trefler 2018) in the case of global trade, the introduction
of improved MT on eBay is an early opportunity
to assess the economic effects of AI using
plausible natural experiments.

1.1.2. Language Barriers in Trade. Empirical studies
using gravity models, which are formally derived in
Anderson and Van Wincoop (2003), have established
a robust negative correlation between bilateral trade
and language barriers. Typically, researchers regress
bilateral trade on a “common language” dummy and
find that this coefficient is strongly positive (Egger
and Lassmann 2012).3 However, these cross-sectional
regressions are vulnerable to endogeneity biases even
after controlling for the usual set of variables in the
gravity equation. For example, two countries with the
same official language (e.g., the United Kingdom and
Australia) can also be similar in preferences for food,
clothing, entertainment, and so forth. Without exogenous
variation in one or the other, it is impossible to
tease out the language effect on trade.
Our paper exploits a natural experiment on eBay
that provides exactly such an exogenous change,
namely a large reduction in the language barrier, and
assesses its effect on international trade. The online
marketplace provides us with a powerful laboratory
to study the consequences on bilateral trade after this
decrease in language barriers for a given language pair.
Our finding that a quality upgrade of machine translation
could increase exports by about 10.9% is consistent
with Lohmann (2011) and Molnar (2013), who
argue that language barriers may be far more trade
hindering than previously suggested.

1.1.3. Peer-to-Peer Platforms and Matching Frictions.
Einav et al. (2016) and Goldfarb and Tucker (2017)
provide great surveys on how digital technology has
reduced matching frictions and improved market efficiency.
Reduced matching frictions affect price dispersion
as evidenced in Brynjolfsson and Smith (2000),
Brown and Goolsbee (2002), Overby and Forman
(2014), and Cavallo (2017). These reduced frictions
also mitigate geographic inequality in economic activities
in the case of ride-sharing platforms (Lam and Liu
2017, Liu et al. 2018), short-term lodging platforms
(Farronato and Fradkin 2018), crowdfunding platforms
(Catalini and Hui 2017), and e-commerce platforms
(Blum and Goldfarb 2006, Lendle et al. 2016, Cowgill
and Dorobantu 2018, Hui 2019). We contribute to this
literature by documenting the significant matching
frictions between consumers and sellers who speak
different languages. Specifically, we find that efforts
to remove language barriers increase market efficiency
substantially.


[...]

2. Background

[...]

Prior to eMT, eBay used Bing Translator for query
and itemtitle translation. Therefore, the policy treatment
here is an improvement in translation quality. To understand
the magnitude of quality improvement, we
follow the MT evaluation literature and report qualities
based on both the BLEU score and human evaluation.
The BLEU score is an automated measure that has
been shown to highly correlate with human judgment
of quality (Callison-Burch et al. 2006). However, BLEU
scores are not easily interpretable and should not be
compared across languages (Denkowski and Lavie
2014). Generally, scores over 30 reflect understandable
translations, and scores over 50 reflect good
translation (Lavie 2010). On the other hand, although
human evaluations are highly interpretable, they are
very costly and can be less consistent.

A comparison of Bing and eMT translation for item
titles from English into Spanish revealed the BLEU
score increased from 41.01 to 45.24, and human acceptance
rate (HAR) increased from 82.4% to 90.2%.
To compute HAR, three linguistic experts vote either
yes or no for translations based on adequacy only
(whether the translation is acceptable for minimum
understanding), and the majority vote is then used to
determine the translation quality. In comparison, the
BLEU score is rated based on both adequacy and fluency
because it compares the MT output with human translation.
Therefore, in cases in which the grammar and
style of translation are not of first-order importance,
such as in listing titles, one might prefer using HAR for
measuring translation quality.

From 2018... Estimates fail to associate gender equality measures with gender segregation in higher education; religiosity is significantly associated with lower gender segregation in higher education

Graduates’ opium? Cultural values, religiosity and gender segregation by field of study. Izaskun Zuazu. Jun 2018. http://www.iaffe.org/media/cms_page_media/788/Izaskun_Zuazu.pdf

Abstract: This paper studies the relationship between cultural values and gender distribution across fieldsof study in higher education. I compute national, field and subfield-level gender segregation indicesfor a panel dataset of 26 OECD countries for 1998-2012. This panel dataset expands the focus ofprevious macro-level research by exploiting data on gender segregation in specific subfields of study. Iconsider two focal cultural traits: gender equality and religiosity, and control for potential segregationfactors, such as labour market and educational institutions, and aggregate-level gender disparities inmath performance and beliefs among young people. The estimates fail to associate gender equality measures with gender segregation in higher education. Religiosity is significantly associated with lower gender segregation in higher education. However, gender gaps in math beliefs seem to be stronger predictors of national-level gender segregation. Field and subfield-level analyses reveal that religiosity is associated with less gender-segregated fields of education, science, and health, and specifically with the subfield of social services.

Keywords: horizontal gender segregation, higher education, cultural values, religiosity, math beliefs,association index.
JEL:A13, I24, J16

6 Conclusion

Persisting levels of gender segregation across fields of study in Western countries seem at odds with the
increase in female participation in higher education. This observation is particularly puzzling against the
backdrop of a rmative action, anti-discrimination policies, and gender-egalitarian ideals in developed
countries. The literature highlights individual factors (gender gaps in preferences and foreseeing family
obligations) and external factors (economic structure, institutions, discrimination) as causes of gender
segregation. This paper studies whether cultural values, in particular gender equality and religion, play
a role in horizontal gender segregation in higher education.

I construct a panel dataset with information on gender segregation indices at national level, at 9-field level
and at 23-subfield level for 26 OECD countries for 1998-2012. I link this data with two focal cultural traits:
Gender equality, measured alternatively on the basis of either the Gender Inequality Index (UNDP) or the
Gender Equality measure (IDEA), and religiosity, taken from the World Value Survey. I propose fixed-effects
models that control for potential segregative factors such as economic structural change, labour
market and educational systems features. The estimates fail to associate gender (in)equality measures
with a significant role in horizontal gender segregation. By contrast, religiosity is significantly associated
with lower levels of horizontal gender segregation.

I expand the models seeking to control for gender gaps in math beliefs developed during the youthhood.
Using two waves of data taken from PISA surveys, I find a contemporaneous association between gender
gaps in anxiety, self-concept and self-e cacy with higher gender segregation of graduates across fields of
study. These gaps seem to be stronger predictors of national-level gender segregation than religiosity.
Field and subfield-levels analyses pinpoint to a robust association between religiosity and lower segregation
levels in the fields of agriculture and health and welfare, and more specifically in the subfield of social
services.

From a policy viewpoint, the role of religiosity may be controversial. However, the findings regarding
gender gaps in math beliefs tend to indicate that e orts to close gaps between boys and girls might
enhance a more gender-equal distribution across fields of study in higher education. Nevertheless, it
should be stressed that the findings above are based on macro-level data on segregation, and should be
taken with caution. Two natural ways to extend this paper would be first to scrutinize whether there is
any link between cultural traits and vertical segregation, i.e. gender segregation at di erent attainment
levels within higher education; and second to expand the gender gaps in ability perceptions among young
people into other dimensions, such as reading and science.

As expected, exploitation had concurrent and longitudinal associations with bullying, but unexpectedly empathic concern only had concurrent associations and no longitudinal associations with bullying

Empathy, Exploitation, and Adolescent Bullying Perpetration: a Longitudinal Social-Ecological Investigation. Ann H. Farrell, Anthony A. Volk, Tracy Vaillancourt. Journal of Psychopathology and Behavioral Assessment, December 4 2019. https://link.springer.com/article/10.1007/s10862-019-09767-6

Abstract: Empathy has been often negatively associated with bullying perpetration, whereas tendencies to be exploitative have been relatively understudied with bullying. Empathic concern and exploitation may also indirectly link distal social-ecological factors to bullying perpetration. Therefore, the associations among personality (i.e., empathic concern, exploitation), self-perceived social-ecological factors (school bonding, social resources), and bullying perpetration were examined in a sample of 531 adolescents across three years of high school in Ontario, Canada (i.e., Grades 9 to 11; mean age 14.96 [SD = 0.37] in Grade 9). As expected, exploitation had concurrent and longitudinal associations with bullying, but unexpectedly empathic concern only had concurrent associations and no longitudinal associations with bullying. Also as expected, exploitation indirectly linked self-perceived social resources to bullying perpetration, but unexpectedly there were no indirect effects with empathic concern. Findings suggest a complex social ecology whereby a lack of empathic concern may remain an important correlate of bullying within each year of high school, whereas exploitative tendencies may be an important predictor of bullying across the high school years, including to strategically leverage self-perceived social resources.

Keywords: Bullying Adolescents Exploitation Empathic concern Social-ecology

From the main author's PhD Thesis, excerpts of the general discussion (chp 5):

Increasing evidence supports the suggestion that adolescence may be a
developmental period when bullying can be adaptively used to acquire material, social,
and romantic resources (Volk, Dane, & Marini, 2014). Bullying may be adaptive under a
specific combination of proximate intrinsic and distal extrinsic social ecological factors.
In particular, genetically influenced personality traits may indirectly link broader
environments to adolescent bullying. The purpose of this dissertation was to investigate
the associations between exploitative personality traits and broader social ecologies
(family, peers, school, community, and economic) to see how they independently and
indirectly facilitated adolescent bullying perpetration. These associations were examined
concurrently, longitudinally, and experimentally in three studies. My prediction that the
broader social environments would filter through exploitative personality traits to
indirectly associate with bullying perpetration was largely supported throughout these
three studies.
In Study 1, I found that environmental variables from three different ecological
systems (micro-, meso-, and macro-) were concurrently associated to both direct (i.e.,
physical, verbal) and indirect (i.e., social) forms of adolescent bullying primarily through
a trait capturing exploitation (i.e., lower Honesty-Humility). Direct bullying also had
indirect associations from social ecological variables through a trait capturing
recklessness (i.e., lower Conscientiousness). To extend on Study 1, I examined
personality-environment associations in a sample of adolescents longitudinally. I found
that exploitation, but not empathy, was longitudinally associated with bullying
perpetration across the first three years of high school. Additionally, social ecological
variables, in particular social status and family functioning, were longitudinally
associated with exploitation, and social status was indirectly longitudinally associated
with bullying through exploitation. Finally, given that Studies 1 and 2 were correlational,
in Study 3, I examined whether bullying perpetration could be simulated through point
allocations in economic games in a laboratory setting. I found that economic games can
be a novel way to experimentally investigate bullying perpetration. Self-report bullying
and selfish Dictator Game point allocations were both related to one another and an
exploitative personality trait (i.e., lower Honesty-Humility). Also, the association
between the environment and both forms of behavior were indirectly facilitated through
this exploitative trait. These three studies together contributed two overall themes in the
social ecology of adolescent bullying perpetration. First, these studies demonstrated the
significance of the role of exploitative personality traits, as opposed to a lack of empathy,
general disagreeableness, or impulsivity, within the context of adaptive adolescent
bullying. Second, these three studies demonstrated a complex social ecology of bullying,
whereby broader social environments from multiple ecological systems can indirectly
facilitate bullying perpetration through exploitative personality traits.


Antisocial Personality and Bullying Perpetration: The Importance of Exploitation

Across all three studies, it was evident that traits capturing exploitation were the
most prominent personality correlates of adolescent bullying perpetration. In both Studies
1 and 3, lower Honesty-Humility was significantly associated with higher bullying
perpetration and selfish Dictator Game point allocations (i.e., an experimental proxy for
bullying). In Study 2, higher exploitation was longitudinally associated with bullying
perpetration. These results are consistent with previous concurrent associations between
adolescent bullying and Honesty-Humility (e.g., Book, Volk, & Hosker, 2012; Farrell,
Della Cioppa, Volk, & Book, 2014), experimental studies on economic game behavior
and Honesty-Humility (e.g., Hilbig, Thielmann, Hepp, Klein, & Zettler, 2015; Hilbig,
Thielman, Wührl, & Zettler, 2015; Hilbig & Zettler, 2009; Hilbig, Zettler, Leist, &
Heydasch, 2013), and finally longitudinal studies on bullying perpetration and narcissism
(i.e., comprised of exploitation and self-superiority; Fanti & Henrich, 2015). It was
evident that adolescents may be strategically exploiting weaker and vulnerable peers to
maximize self-gain, while minimizing costs like victim retaliation. More importantly, my
results demonstrate that a predatory, exploitative tendency may be the most relevant
personality risk factor for engaging in bullying, over and above other personality traits
related to antisocial tendencies.
In contrast to previous studies that found bullying perpetration is often associated
with personality traits such as a lack of empathy, a general tendency to be disagreeable or
angry, and higher impulsivity (e.g., Bollmer, Harris, & Milich, 2006; Caravita, Di Blasio,
& Salmivalli, 2009; Tani, Greeman, Schneider, & Fregoso, 2003), I found lower
Honesty-Humility and higher exploitativeness were associated with bullying, despite
controlling for these other antisocial personality traits. In both Studies 1 and 3, I found
that lower Honesty-Humility was the strongest correlate of bullying perpetration, over
and above the other HEXACO personality traits. Although indirect and direct forms of
bullying and Dictator Game point allocations were both negatively related with lower
Agreeableness (and in Study 1 additionally related to lower Conscientiousness), Honesty-
Humility was the strongest correlate. Thus, it appears that predatory exploitation over
weaker individuals may be the driving personality factor facilitating bullying, even if the
other antisocial traits are still important and associated with bullying. These results are
consistent with recent findings that although Honesty-Humility, Emotionality,
Agreeableness, and Conscientiousness from the HEXACO were all associated with
antisocial tendencies, Honesty-Humility was the largest and driving contributor of
antisociality (Book et al., 2016; Book, Visser & Volk, 2015; Hodson et al., 2018).
However, it is important to note that lower Conscientiousness was additionally a
significant multivariate predictor of direct, but not indirect, bullying. This result
demonstrates that in addition to reflecting strategical exploitation, direct forms of
bullying like hitting, pushing, or kicking, may reflect a risky form of antisocial behavior
that is associated with a general recklessness (Volk et al., 2014). These indirect
associations are also consistent with theories of a faster life history, which posit that
certain individuals who experience competitive or adverse social environments may be
more likely to engage in more impulsive and aggressive behavior to obtain immediate,
short-term access to resources, and bullying may be one behavior that can reflect this
strategy (Dane, Marini, Volk, & Vaillancourt, 2017; Del Giudice & Belsky, 2011;
Hawley, 2011). Interestingly, unlike Agreeableness and Conscientiousness, which had at
least univariate associations with bullying, lower Emotionality or lower empathy had the
fewest univariate and multivariate associations with adolescent bullying.
Although contrasting with the prevalent theories linking lower empathy with
bullying, our lack of association agrees with more contemporary theories of adolescent
bullying as an adaptive, predatory strategy. In Study 1, lower Emotionality was not
significantly related at either the univariate or multivariate levels with bullying, and in
Study 2, lower empathy was only concurrently, but not longitudinally, related with higher
bullying. My results are in contrast to those of previous researchers who found significant
associations between child bullying and lower empathy (e.g., Caravita et al., 2009; Zych,
Ttofi, & Farrington, 2016). Instead, my findings support my prediction that instead of a
lack of emotional recognition or response, a predatory exploitation of others’ weaknesses
may be an important reason why adolescents bully. This may be one potential reason why
empathy related interventions may have been largely ineffective for adolescents (Yeager,
Fong, Lee, & Espelage, 2015). Taken together, all three studies not only support existing
literature on the concurrent association between exploitative style traits and adolescent
bullying (e.g., Book et al., 2012), but extend on these findings by providing both quasiexperimental
and longitudinal evidence for this association. These results with
personality and bullying also suggest that not every risk factor for bullying affects all
adolescents in the same way. Instead, adolescents with specific personality traits may be
more likely and willing to use bullying. Further, adolescents with these personality traits
can respond to, or are influenced by, particular environments in multiple ways (Caspi et
al., 2002; Marceau et al., 2013; Moffitt, 2005; Scarr & McCartney, 1983). In my thesis,
these associations between environment and personality were evident through the
multiple social environmental variables that were indirectly associated with bullying
through exploitative personality styles.


Bullying Perpetration and Indirect Associations with Broader Social Ecology

Across the three studies, it was evident that not all social environments facilitate
bullying in the same way for all adolescents. Instead as expected, I found that multiple
adverse and risky social environment variables filtered specifically through exploitative
personality traits to indirectly facilitate adolescent bullying. These social environmental
variables were from multiple ecological systems ranging from proximate economic
power contexts and peer and family relationships, to distal school and community
variables. Starting with the more proximate factors, all three studies demonstrated that
social relationships in the microsystem (i.e., immediate social context), had indirect
associations with bullying most frequently through either lower Honesty-Humility (i.e.,
Study 1 and 3), or higher exploitation (i.e., Study 2), as opposed to other antisocial
personality traits. Occasionally in Study 1, a proximate ecological factor was found to
have an indirect effect with bullying primarily through Honesty-Humility and secondarily
through lower Conscientiousness. In these instances, the strength of the indirect
associations through Honesty-Humility and Conscientiousness were often comparable, as
indicated through the standardized beta coefficients. The associations with Honesty-
Humility and exploitation may be a result of predatory individuals being able to
strategically take advantage of adverse and/or risky social ecological circumstances.
Adverse social relationships including poorer family dynamics and higher peer
problems, along with powerful social positions such as higher social status (i.e., Study 2)
or higher interpersonal influence (i.e., Study 1 and 3), appeared to be risk factors
indirectly associated with bullying through exploitative personality styles. Individuals
with higher social status or social influence, and individuals who experience negative
social relationships characterized by conflict, lower support and lower warmth, may
exploit these social environments. For example, adolescents who know their parents do
not have much knowledge or care for their whereabouts may take advantage of this lack
of interest by engaging in bullying, knowing that they would have fewer repercussions.
Likewise, adolescents who know they have poorer friendships may exploit these friends
and employ these friends in bullying strategies. Finally, exploiting these relationships
may be especially advantageous for adolescents who have higher social status, as they
would likely have greater influence when navigating their peer networks to effectively
assert their power through bullying tactics. These concurrent results held longitudinally
across three years of adolescence, and also held when manipulating dyadic economic
contexts in a laboratory setting.
These results are consistent with broader evolutionary frameworks that help
explain the use of aggressive behavior. Bullying may be a facultative or conditional
adaptation that an adolescent may consciously or subconsciously decide to engage in
after evaluating his or her own personality traits (i.e., exploitative tendencies; Buss, 2011)
in combination with their broader environments (e.g., friendships, family relationships,
social status; Dane et al., 2017; Del Giudice & Belsky, 2011; Hawley, 2011; Volk,
Camilleri, Dane, & Marini, 2012). Adverse and negative social environments may also
facilitate faster life history strategies that encourage aggressive behavior like bullying (as
opposed to cooperative, long-term strategies) as an immediate means for resources (Del
Giudice & Belsky, 2011; Hawley, 2011). After these assessments of the self and
environment, an adolescent may anticipate the immediate benefits of bullying over
weaker peers may outweigh the costs. Additionally, if previous uses of bullying have
been successful, these dominant and exploitative adolescents may be more inclined to use
this behavior again (Dawkins, 1989). Alternatively, adolescents who possess certain
genetically based personality traits such as exploitative tendencies, may already be more
likely to use coercive or bullying behavior, as opposed to prosocial or cooperative forms
of behavior (Del Giudice & Belsky, 2011).
In addition to evolutionary frameworks, my results are consistent with
Bronfenbrenner’s Ecological Systems Theory (EST; Bronfenbrenner, 1979), and with
recent findings that multiple ecological levels can differentially facilitate bullying
perpetration (e.g., Hong & Espelage, 2012). Furthermore, my results demonstrate that
multiple ecological contexts can have indirect associations with individual differences in
personality, similar to previous ecological studies on bullying (e.g., Barboza et al., 2009;
Lee, 2011; Low & Espelage, 2014). However, my results provide some key novel
contributions. My findings demonstrate that there are indirect associations from adverse
parental and peer relationships and socially powerful positions to bullying, specifically
through exploitative traits, as opposed to other antisocial personality traits. These results
are likely due to the reason that exploitative adolescents may be more willing and able to
take advantage of adverse relationships and powerful positions.
It is likely that within these environmental contexts, exploitative adolescents may
experience more social benefits when using bullying (e.g., increased social status), and
may simultaneously have fewer costs imposed by parents and/or weaker peers. One of the
most noteworthy and prominent social ecological variables that emerged were status
related variables that indicate higher power. Across all three studies, it was evident that
higher social status (i.e., Study 2) or higher interpersonal influence (i.e., Studies 1 and 3)
were commonly associated with bullying through exploitative personality, and this
association held concurrently, longitudinally, and in a laboratory based experimental
setting. Bullying is fundamentally about a power imbalance (Volk et al., 2014). By
definition, bullying requires an individual with more power to inflict harm on a weaker
individual. As evident throughout animal studies (e.g., hierarchy in hyenas; Stewart,
1987), and in research in which human participants are assigned a powerful role (e.g.,
role of prison guards; Haney, Banks, & Zimbardo, 1973), a position of higher status and
social power can be translated into gaining resources at the expense of others. It is not
surprising then, that this fundamental feature that distinguishes bullying from other forms
of aggression is reflected in the broader adolescent social ecology. Those willing to use
power to inflict harm on weaker peers may be more effective in doing so if they have
exploitative, predatory tendencies (as opposed a general lack of concern or empathy for
others). These exploitative tendencies will ultimately assist in taking advantage of higher
social status and influence to strategically bully weaker peers, who are less likely to
defend themselves and/or retaliate.
My findings are similar to previous associations between strategic adolescent
bullying and higher perceived popularity, social status, and influence (Dijkstra,
Lindenberg, & Veenstra, 2008; Garandeau, Lee, & Salmivalli, 2013; Pellegrini & Long,
2002; Reijntjes et al., 2013; Sentse, Veenstra, Kiuru, & Salmivalli, 2015; Sijtsema,
Veenstra, Lindenberg, & Salmivalli, 2009). Additionally, these results are consistent with
those of previous researchers who found that although adolescents who bully can be high
in peer-perceived popularity, power, and status, they are not necessarily socially preferred
or liked by peers as friends (Vaillancourt & Hymel 2006). Researchers have found that
early adolescence is a developmental period when peer-perceived popularity is most
valued (LaFontana & Cillessen, 2010). As a result, given that adolescents may be
exploiting their social status to engage in bullying, my results support the notion that
bullying can be used selectively and adaptively by adolescents for status, a goal that is
highly salient during adolescence. A similar pattern of indirect effects also emerged with
more distal ecological variables.
Adverse mesosystem variables (i.e., interactions among immediate social
environments), and macrosystem variables (i.e., broader cultural attitudes and values)
also indirectly facilitated bullying through either lower Honesty-Humility or higher
exploitation. Risky, negative aspects of the social environment such as higher
neighborhood violence, higher school competition, and adverse school climates were
indirectly associated with bullying. It appears that in addition to immediate social
environments, exploitative adolescents may take advantage of wider negative climates to
engage in bullying for self-gain. These broader adverse environments may not provide
the social structures including discipline that could prevent adolescents from acting on
their exploitative motivations. Thus, in addition to assessments of the self and
environment, adolescents may have learned the benefits of bullying within these
environments outweigh the costs through vicarious reinforcement, consistent with the
Social Learning Theory (Bandura, 1978). The fact that risky environments filtered
through both lower Honesty-Humility and lower Conscientiousness for direct bullying
behavior in Study 1 suggests that while all forms of bullying can be strategically
implemented within the right conditions, direct forms of bullying behavior also reflect a
recklessness for consequences (Volk et al., 2014), and a tendency to engage in riskier,
aggressive behavior for immediate gain (Del Giudice & Belsky, 2011). Accordingly,
these findings further provide support that not all environments affect all adolescents the
same way. Although predatory, exploitative tendencies appear to filter both proximate
and distal adverse social environments for bullying perpetration, there are even subtle
differences in the bullying behavior used. Poorer social relationships, higher social status,
and more competitive and violent school and neighborhood variables appear to be risk
factors for engaging in bullying as a whole. These variables appear to be risk factors for
exploitative adolescents who may strategically take advantage of these contexts to
adaptively bully. However, these adverse environments may also be risk factors for
generally impulsive or reckless adolescents willing to engage in direct forms of bullying.
Accordingly, there appears to be a successful facultative translation of these risky social
environments into adaptive bullying behavior by adolescents with a primarily predatory,
exploitatively personality style.
Taken together, my results are consistent with previous findings on poorer social
relations interacting with Honesty-Humility to predict bullying (e.g., lower parental
knowledge; Farrell, Provenzano, Dane, Marini, & Volk, 2017), and with additional
ecological findings on personality interacting with or indirectly linking social
environments for bullying (Barboza et al., 2009; Lee, 2011; Low & Espelage, 2014). My
findings suggest that adolescents who bully may not necessarily be generally
disagreeable, antisocial individuals with a lack of empathy. Instead, adolescent
perpetrators may be strategic, exploitative individuals who are able to take advantage of
their broader social environments and immediate social influence to gain more benefits,
while simultaneously reducing costs. Both Studies 1 and 2 demonstrated that the distal
and proximate environmental contexts may adaptively filter through an exploitative
personality trait to predict bullying, a behavior rooted in taking advantage of power.
However, Study 3 extended findings from the previous two studies by demonstrating how
proximate contextual factors like power can be manipulated to examine bullying and/or
similarly related competitive behavior, and how these forms of behavior relate with
personality. Despite these significant contributions, this dissertation was not without
limitations.

Induced Mate Abundance Increases Women’s Expectations for Engagement Ring Size and Cost

Induced Mate Abundance Increases Women’s Expectations for Engagement Ring Size and Cost. Ashley Locke, Jessica Desrochers, Steven Arnocky. Evolutionary Psychological Science, December 4 2019. https://link.springer.com/article/10.1007/s40806-019-00214-z

Abstract: Research on some non-human species suggests that an abundance of reproductively viable males relative to females can increase female choosiness and preferences for longer-term mating and resource investment by males. Yet little research has explored the potential influence of mate availability upon women’s preferences for signals of men’s commitment and resource provisioning. Using an experimental mate availability priming paradigm, the present study examined whether women (N = 205) primed with either mate scarcity or abundance would differ in their expectations for engagement ring size and cost. Results demonstrated that women who were primed with the belief that good-quality mates are abundant in the population reported expecting a statistically-significantly larger and more expensive engagement ring relative to women primed with mate scarcity. Results suggest that women flexibly attune their expectations for signals of men’s investment based, in part, upon their perception of the availability of viable mates.

Keywords: Priming Sex ratio Engagement rings Social psychology Evolutionary psychology Mating behavior

---
Sample
205 undergraduate women aged 17 to 39 (M 20 SD 2 87).

Demographic Measures
Prior to the priming task, participants completed measures of age and romantic relationship status (“Are you currently in a committed heterosexual romantic relationship?”)

Mate Availability Priming Task
Using a set of fictitious magazine articles developed by Spielmann, MacDonald, and Wilson 2009 participants were primed with the belief that potential mates were either abundant or scarce. In this task, participants read one of two articles In the mate abundant condition, the article explained the task of finding a new romantic partner as being relatively easy, with the mating population
consisting of many available mates. Conversely, in the mate scarcity condition, the article highlighted the difficulty of finding a new romantic partner, with desirable mates being a scarce resource.

Manipulation Check
Participants then responded to the following two items asking about their own perceptions of mate availability: (1) “It scares me to think there might not be anyone out there for me” and (2) "I feel it is close to being too late for me to find love in my life"

Engagement Ring Preferences
Following Cloud and Taylor 2018 female participants were asked ,,“if this man were to propose to you after an extended period of dating, what is the smallest size engagement ring that you would be satisfied with him giving to you To make their decision, participants saw five identical engagement rings that differed only by carat weight and cost, ranging from 0 50 carats 500 to 1 50 carats 9000 and their selection was recorded (see Fig 1 from Cloud and Taylor 2008).

Are Sex Differences in Mating Preferences Really “Overrated”? The Effects of Sex and Relationship Orientation on Long-Term and Short-Term Mate Preference

Are Sex Differences in Mating Preferences Really “Overrated”? The Effects of Sex and Relationship Orientation on Long-Term and Short-Term Mate Preference. Sascha Schwarz, Lisa Klümper, Manfred Hassebrauck. Evolutionary Psychological Science, December 4 2019. https://link.springer.com/article/10.1007/s40806-019-00223-y

Abstract: Sex differences in mating-relevant attitudes and behaviors are well established in the literature and seem to be robust throughout decades and cultures. However, recent research claimed that sex differences are “overrated”, and individual differences in mating strategies (beyond sex) are more important than sex differences. In our current research, we explore between-sex as well as within-sex differences; further we distinguish between short-term and long-term relationship orientation and their interactions with sex for predicting mate preferences. In Study 1, we analyzed a large dataset (n = 21,245) on long-term mate characteristics. In Study 2 (n = 283), participants indicated their preference for long-term as well as short-term partners. The results demonstrate the necessity to include both intersexual as well as intrasexual differences in mating strategies. Our results question the claim that sex differences in mate preferences are “overrated.”

Keywords: Sex differences Mate preferences Sociosexual orientation Long-term relationship orientation Short-term relationship orientation Online dating

Education's marginal cognitive benefit does not reach a plateau until 17 years of education; those with low childhood intelligence derive the largest benefit of education

The influence of educational attainment on intelligence. Emilie Rune Hegelund et al. Intelligence. Volume 78, January–February 2020, 101419. https://doi.org/10.1016/j.intell.2019.101419

Highlights
•    Education has a positive influence on intelligence.
•    The marginal cognitive benefit of education does not reach a plateau until 17 years of education.
•    Individuals with low childhood intelligence derive the largest benefit from education.
•    Findings of relatively small cognitive benefits might be explained by selection bias.

Abstract: Education has been found to have a positive influence on intelligence, but to be able to inform policy, it is important to analyse whether the observed association depends on the educational duration and intelligence prior to variations in educational attainment. Therefore, a longitudinal cohort study was conducted of all members of the Metropolit 1953 Danish Male Birth Cohort who were intelligence tested at age 12 and appeared before the Danish draft boards (N = 7389). A subpopulation also participated in the Copenhagen Aging and Midlife Biobank (N = 1901). The associations of educational attainment with intelligence in young adulthood and midlife were estimated by use of general linear regression with adjustment for intelligence test score at age 12 and family socioeconomic position. Results showed a positive association of educational attainment with intelligence test scores in both young adulthood and midlife after prior intelligence had been taken into account. The marginal cognitive benefits depended on the educational duration but did not reach a plateau until 17 years. Further, intelligence test score at age 12 was found to modify the association, suggesting that individuals with low intelligence in childhood derive the largest benefit from education. Comparing the strength of the association observed among participants and non-participants in our midlife study, we showed that selection due to loss to follow-up might bias the investigated association towards the null. This might explain previous studies' findings of relatively small cognitive benefits. In conclusion, education seems to constitute a promising method for raising intelligence, especially among the least advantaged individuals.


4.2. Comparison with the existing literature
The finding of a positive association between educational attainment
and intelligence test scores after prior intelligence has been taken
into account is consistent with extant literature (Clouston et al., 2012;
Falch & Massih, 2011; Ritchie, Bates, Der, Starr, & Deary, 2013), including
a recent meta-analysis (Ritchie & Tucker-Drob, 2018). More
specifically, our results suggested an average increase in intelligence
test score of 4.3 IQ points per year of education in young adulthood and
1.3 IQ points per year of education in midlife. The effect estimate in
young adulthood is considerably higher than the effect estimate of 1.2
IQ points (95% confidence interval: 0.8, 1.6) reported in the metaanalysis
for the control prior intelligence design. However, in a simultaneous
multiple-moderator analysis, the authors report an adjusted
effect estimate of 2.1 IQ points (95% confidence interval: 0.8, 3.4),
taking into account the possible influence of age at early test, age at
outcome test, outcome test category, and male-only studies. Besides
these possible moderators, contextual factors might also account for the
contrasting findings between our study and the seven studies included
in the meta-analysis. However, it is important to notice that our effect
estimate in midlife is consistent with the findings of the meta-analysis,
suggesting that sample selectivity might have influenced the association
observed in midlife in our study and perhaps the associations observed
in the cohort studies included in the meta-analysis as well. This is
supported by findings of higher educational attainment among the individuals
in our study population who participated in the midlife study,
higher IQ at age 12 (IQ mean: 103.2 vs. 98.9) as well as our finding of
effect measure modification. If both educational attainment and intelligence
are positively associated with participation in studies and
individuals with low intelligence in childhood derive the largest benefit
from education, selection will bias the investigated associations towards
the null. This is probably the reason why the study with the least
selected sample in the meta-analysis is the one reporting the largest
effect estimate (Falch & Massih, 2011). Based on a male population who
were initially intelligence tested in school at age 10 and later at the
mandatory military draft board examination at age 20, the authors
found an increase in intelligence test score of 3.5 IQ points per year of
education (95% confidence interval: 3.0, 3.9), which is much more in
line with our results.
Our finding of a stronger association between educational attainment
and intelligence test scores in young adulthood compared with
midlife in the subpopulation of individuals who participated in both
examinations is consistent with the findings of the meta-analysis
(Ritchie & Tucker-Drob, 2018). One possible explanation for this
finding might be that schooling has a larger influence on intelligence
compared with vocational education or training, which mainly takes
place after the age of 18. However, the finding might also be explained
by the smaller time gap between the measurements of educational attainment
and intelligence, as the measurement of educational attainment
in midlife in most cases will reflect one's educational attainment
before the age of 30. The older the age at outcome testing, the larger the
time gap and thus more additional factors might influence the association,
such as the individual's occupational complexity and health
(Smart, Gow, & Deary, 2014; Waldstein, 2000).
In general, our finding of a positive association between educational
attainment and intelligence test scores should be interpreted with
caution. As previously written, it is difficult to separate the positive
influence of educational attainment on intelligence from the influence
of selection by prior intelligence, whereby individuals with higher intelligence
test scores prior to variations in educational attainment
progress further in the educational system. Although our results show a
strong positive association between educational attainment and intelligence
test scores after prior intelligence has been taken into account,
a hierarchical analysis of our data suggests that educational attainment
only increases the amount of explained variance in later IQ by
7% when IQ at age 12 is already accounted for (R2 = 0.46 vs
R2 = 0.53; p < .001). Therefore, our findings most likely not only
reflect the positive influence of educational attainment on intelligence,
but also a residual influence of selection processes which our statistical
analyses were not able to take into account.
To answer one of our specific aims, we also investigated whether the
increase in intelligence test score for each extra year of education depended
on the educational duration. Irrespective of whether intelligence
was measured in young adulthood or midlife, intelligence test
scores were found to increase with increasing years of education in a
cubic relation, suggesting that the increase in intelligence test score for
each extra year of education diminishes with increasing length of
education. This finding supports the hypothesis proposed by the authors
of a recent study, who, based on their own data and the existing literature,
suggest that the influence of educational attainment on intelligence
eventually might reach a plateau (Kremen et al., 2019).
However, where the authors of the previous study suggest that this
plateau might already be reached by late adolescence as their findings
show no significant association between educational attainment and
intelligence test score in midlife after IQ at age 20 has been taken into
account, we find no plateau until approximately 17 years of education.
In fact, replicating the previous study's statistical analyses we find an
average increase in intelligence test score in midlife of 0.8 IQ points per
year of education taking IQ at age 20 into account (Supplementary
Table 4). We speculate whether this contrasting finding might be explained
by the representativeness of the study populations as the previous
study is based on a selected sample of twins serving in the
American military at some point between 1965 and 1975. Thus, a study
based on the two Lothian birth cohorts, which like our study includes a
follow-up examination of a population-representative survey in childhood,
finds a weighted average increase in intelligence test score in late
life of 1.2 IQ points per year of education taking IQ at age 11 into
account (Ritchie et al., 2013). However, the contrasting finding might
also be explained by residual confounding due to the use of non-identical
baseline and outcome intelligence tests in our study. Nevertheless,
in our study, the strongest association between educational attainment
and intelligence in midlife was observed in upper-secondary school, i.e.
around 10–13 years of education. A possible explanation for this finding
might be that pupils up to and including upper-secondary school receive
general education, which improves exactly what the intelligence
tests included in our study most likely measure: General cognitive
ability. After upper-secondary school, individuals start to specialize in
different fields, which may explain why the increase in intelligence test
score for each extra year of education diminishes. However, the cubic
tendency was relatively weak and as written above the association
between educational attainment and intelligence did not reach a plateau
until approximately 17 years of education, corresponding to the
completion of a Master's degree program. As our study is the first to
investigate whether the increase in intelligence test score for each extra
year of education depends on the educational duration, replication of
our finding is needed – preferably in studies with access to the same
baseline and outcome test.
Finally, to answer another of our specific aims, we investigated
whether the increase in intelligence test score for each extra year of
education depended on the intelligence prior to variations in educational
attainment. The results showed that the increase in intelligence
test score for each extra year of education was higher in the group of
individuals with an IQ < 90 compared with the group of individuals
with an IQ of 90–109. Although this finding clearly needs to be replicated,
it is in line with the findings of a Danish study investigating
whether distributional changes accompanied the secular increases in
intelligence test scores among males born in 1939–1958 and
1967–1969 (Teasdale & Owen, 1989). According to the authors of that
study, a possible explanation of why individuals with low intelligence
in childhood derive the largest benefit from education is that the Danish
school system for the last seven decades mainly has focused on improving
the abilities of the least able (Teasdale & Owen, 1989).
Therefore, future studies are needed to investigate whether our finding
is peculiar to the Danish school system or whether it can be generalized
to school systems in other countries.

Self-reported ratings of deception ability were positively correlated with telling a higher frequency of white lies & exaggerations, & telling the majority of lies to colleagues, friends, & romantic partners

Verigin BL, Meijer EH, Bogaard G, Vrij A (2019) Lie prevalence, lie characteristics and strategies of self-reported good liars. PLoS ONE 14(12): e0225566. https://doi.org/10.1371/journal.pone.0225566

Abstract: Meta-analytic findings indicate that the success of unmasking a deceptive interaction relies more on the performance of the liar than on that of the lie detector. Despite this finding, the lie characteristics and strategies of deception that enable good liars to evade detection are largely unknown. We conducted a survey (n = 194) to explore the association between laypeople’s self-reported ability to deceive on the one hand, and their lie prevalence, characteristics, and deception strategies in daily life on the other. Higher self-reported ratings of deception ability were positively correlated with self-reports of telling more lies per day, telling inconsequential lies, lying to colleagues and friends, and communicating lies via face-to-face interactions. We also observed that self-reported good liars highly relied on verbal strategies of deception and they most commonly reported to i) embed their lies into truthful information, ii) keep the statement clear and simple, and iii) provide a plausible account. This study provides a starting point for future research exploring the meta-cognitions and patterns of skilled liars who may be most likely to evade detection.

Discussion
We found that self-reported good liars i) may be responsible for a disproportionate amount of lies in daily life, ii) tend to tell inconsequential lies, mostly to colleagues and friends, and generally via face-to-face interactions, and iii) highly rely on verbal strategies of deception, most commonly reporting to embed their lies into truthful information, and to keep the statement clear, simple and plausible.

Lie prevalence and characteristics
First, we replicated the finding that people lie, on average, once or twice per day, including its skewed distribution. Nearly 40% of all lies were reported by a few prolific liars. Furthermore, higher self-reported ratings of individuals’ deception ability were positively correlated with self-reports of: i) telling a greater number of lies per day, ii) telling a higher frequency of white lies and exaggerations, iii) telling the majority of lies to colleagues and friends or others such as romantic partners, and iv) telling the majority of lies via face-to-face interactions. Importantly, skewed distributions were also observed for the other lie characteristics, suggesting that it may be misleading to draw conclusions from sample means, given that this does not reflect the lying behaviours of the average person. A noteworthy finding is that prolific liars also considered themselves to be good liars.

The finding that individuals who consider themselves good liars report mostly telling inconsequential lies is somewhat surprising. This deviates from the results of a previous study, which showed that prolific liars reported telling significantly more serious lies, as well as more inconsequential lies, compared to everyday liars [15]. However, small, white lies are generally more common [18] and people who believe they can get away with such minor falsehoods may be more inclined to include them frequently in daily interactions. It is also possible that self-reported good liars in our sample had inflated perceptions of their own deception ability because they tell only trivial lies versus lies of serious consequence.

Regarding the other lie characteristics, we found a positive correlation between self-reported deception ability and telling lies to colleagues, friends and others (e.g., romantic partners). This variation suggests that good liars are perhaps less restricted in who they lie to, relative to other liars who tell more lies to casual acquaintances and strangers than to family and friends [22]. Our results also showed that good liars tended to prefer telling lies face-to-face. This fits the findings of one of the only other studies to examine characteristics of self-reported good versus poor liars, which found that self-perceived good liars most commonly lied via face-to-face interactions versus through text chat [42]. This could be a strategic decision to deceive someone to their face, since people may expect more deception via online environments [43]. As researchers continue to examine the nature of lying and to search for ways of effectively detecting deception, it is important to recognize how certain lie characteristics may influence individuals’ detectability as liars.

Deception strategies
We also isolated the lie-telling strategies of self-reported good liars. People who identified as good liars placed a higher value on verbal strategies for successfully deceiving. Additional inspection of the verbal strategies reported by good liars showed that commonly reported strategies were embedding lies into truthful information and keeping their statements clear, simple and plausible. In fact, good liars were more likely than poor liars to endorse using these strategies, as well as matching the amount and type of details in their lies to the truthful part/s of their story, and providing unverifiable details. A common theme among these strategies is the relation to truthful information. This fits with the findings of previous literature, that liars typically aim to provide as much experienced information as possible, to the extent that they do not incriminate themselves [35, 44]. Additionally, good liars used plausibility as a strategy for succeeding with their lies. This reflects the findings of the meta-analysis by Hartwig and Bond [45] that implausibility is one of the most robust correlates of deception judgements, and the results of DePaulo et al. [26] that one of the strongest cues to deception is liars’ tendency to sound less plausible than truth tellers (d = -0.23).

We also found that self-reported poor liars were more likely than good liars to rely on the avoidance strategy (i.e., being intentionally vague or avoiding mentioning certain details). Previous research suggests that this is one of the most common strategies used by guilty suspects during investigative interviews [46]. Additionally, all liars in our study expressed behavioural strategies as being important for deceiving successfully. This could be explained by the widespread misconceptions about the associations between lying and behaviour, for example that gaze aversion, increased movement or sweating are behaviours symptomatic of deception [2, 47].

There was inconsistency in our data between the responses to the qualitative strategy question and the multiple-response strategy question. Based on the qualitative strategy data it seems that Good, Neutral, and Poor liars do not differ in their use of strategies. However, robust differences emerged when we evaluated participants’ endorsement of the predetermined strategies. One explanation for this finding is the difficulty people perceive when they have to verbalize the reasons for their behavior. Ericsson and Simon [48] suggest that inconsistencies can occur especially when the question posed is too vague to elicit the appropriate information, which might have been the case in our study. Another explanation for the discrepancy in the data between the two measurement procedures is that data-driven coding is inherently susceptible to human subjectivity, error, and bias [49, 50]. Such limitations apply to a lesser extent to coding based on predetermined categories that are derived from psychological theory, an approach which has been heavily used within the deception literature [2]. In any case, future research should continue exploring the deception strategies of good liars using a variety of methodological approaches. In particular, it would be beneficial to establish more reliable techniques for measuring interviewees’ processing regarding their deception strategies. One potential idea could be to explore the effectiveness of using a series of cued questions to encourage the recall of specific aspects of interviewees’ memory or cognitive processing. Another suggestion is to combine the data-driven and theory-driven approaches, whereby the coding system is generated inductively from the data but the coders draw from the theoretical literature when identifying categories [50].

Limitations
Some methodological considerations should be addressed. First, the results of the present study are drawn from participants’ self-reports about their patterns of deception in daily life. Sources of error associated with such self-report data limit our ability to draw strong inferences from this study. However, previous research has validated the use of self-report to measure lying prevalence by correlating self-reported lying with other measures of dishonesty [17]. Moreover, self-report data may not be as untrustworthy as critics argue, and in some situations, it may be the most appropriate methodology [51]. This study was intended as an initial examination of the strategies and preferences of good liars, and surveying liars for their own perspectives provided a novel source of insight into their behaviour. A constraint to the generalizability of this research is that we did not establish the ground truth as to whether self-reported good liars are indeed skilled deceivers. Future research could attempt to extend our findings by examining deceivers’ lie frequency, characteristics, and strategies after systematically testing their lie-telling ability within a controlled laboratory setting.

Second, one of the most frequent concerns about using Amazon MTurk relates to low compensation and resulting motivation [52, 53]. We took measures to ensure that our remuneration to participants was above the fair price for comparable experiments. Importantly, data collected through MTurk produces equivalent results as data collected from online and student samples [52, 54–58]. As well, mTurk surveys have been shown to produce a representative sample of the United States population that yields results akin to those observed from more expensive survey techniques, such as telephone surveys [57]. It speaks to the validity of our data, for example, that the self-reported prevalence of lies, and the endorsement of nonverbal deception strategies, replicates previous research. Nonetheless, the results of this study could be advanced if future research i) directly replicates our survey amongst different populations, for instance university students, and ii) conceptually replicates this research by evaluating different methodological approaches for measuring deception ability (e.g., via controlled evaluation) and good liars’ strategies for deceiving (e.g., via cued recall).

Implications and future research
This study explored the deception characteristics and strategies used by self-reported good liars. Deception researchers generally agree that the most diagnostic information is found in the content of liars’ speech [59]. Content-based cues to deception, however, may be less effective for detecting good liars who rely highly on verbal strategies of deception. This could even offer an explanation for the modest effect sizes observed in the deception literature [60]. For instance, good liars in our study reported to strategically embed their lies into truthful information. This finding has potential implications for the reliability of credibility assessment tools that derive from the assumption that truth tellers’ statements are drawn from memory traces whereas liars’ statements are fabricated from imagination [61, 62]. If good liars draw on their memory of truthful previous experiences, then their statements may closely resemble those of their truth telling counterparts. Another interesting observation was that self-reported good liars were more likely than poor liars to provide unverifiable details. This fits with the findings of previous literature on the VA, which contends that liars provide information that cannot be verified to balance their goals of being perceived as cooperative and of minimizing the chances of falsification by investigators [32, 33]. A fruitful avenue of future research could be to further explore liars’ strategic inclusion of truthful information and unverifiable details. Doing so may give lie detectors an advantage for unmasking skilled liars. It would also be interesting for future research to examine how good versus poor liars are affected by certain interview techniques designed to increase the difficulty of lying such as the reverse-order technique [63].

Females are more proactive, males are more reactive: neural basis of the gender-related speed/accuracy trade-off in visuo-motor tasks

Females are more proactive, males are more reactive: neural basis of the gender-related speed/accuracy trade-off in visuo-motor tasks. V. Bianco et al. Brain Structure and Function, December 3 2019. https://link.springer.com/article/10.1007/s00429-019-01998-3

Abstract: In the present study, we investigated neural correlates associated with gender differences in a simple response task (SRT) and in a discriminative response task (DRT) by means of event-related potential (ERP) technique. 120 adults participated in the study, and, based on their sex, were divided into two groups matched for age and education level. Behavioral performance was assessed with computing response speed, accuracy rates and response consistency. Pre- and post-stimulus ERPs were analyzed and compared between groups. Results indicated that males were faster than females in all tasks, while females were more accurate and consistent than males in the more complex tasks. This different behavioral performance was associated with distinctive ERP features. In the preparation phase, males showed smaller prefrontal negativity (pN) and visual negativity (vN), interpreted as reduced cognitive preparation to stimulus occurrence and reduced reliance on sensory proactive readiness, respectively. In the post-stimulus phase, gender differences were present over occipital (P1, N1, P2 components) and prefrontal (pN1, pP1, pP2 components) areas, suggesting allocation of attentional resources at distinct stages of information processing in the two groups. Overall, the present data provide evidence in favor of a more proactive and cautious cognitive processing in females and a more reactive and fast cognitive processing in males. In addition, we confirm that (1) gender is an important variable to be considered in ERP studies on perceptual processing and decision making, and (2) the pre-stimulus component analysis can provide useful information concerning neural correlates of upcoming performance.

Keywords: Gender differences Speed–accuracy trade-off Motor behavior Proactive control Decision making Predictive brain

Conclusions 
Present data suggest that in simple and complex visuo-motor tasks, males and females allocate their cortical resources in diverse ways, possibly leading to the well-documented gender-related speed/accuracy trade-off in visuo-motor per-formance. When the task is very simple, both preparatory (the BP) and reactive (the pP1, P2 and P3) cortical process-ing are enhanced in males with respect to females, leading to faster responses. When the task is more complex (implying stimulus discrimination and response selection), females’ proactive allocation of more cortical resources at both pre-frontal (pN) and sensory (vN) level, as well as several reac-tive stages after stimulus onset (the pN1, the P1, and the P3), leads to relatively slow and very accurate responses. In contrast, males allocate a reduced level of pre-stimulus sustained attention to the task (smaller pN and vN), possi-bly compensating with enhanced reactive attention at visual level processing (larger N1 and P2). Even though the neu-ral processing associated with S–R mapping (the pP2) is generally enhanced in males (for both target and non-target stimuli), signals associated to different stimulus categories are less distinguishable in males than females, as indicated by dpP2 effect, possibly facilitating female accuracy in a complex task.

Present research provides evidence that gender is an important variable to be considered in neurocognitive studies of perceptual decision making; this variable should be considered while planning experimental designs or interpreting the results because, per se, could explain the speed/accuracy trade-off in visuo-motor performance and relative differences in brain functions. In contrast, some studies excluded females from their samples or ignored gender as a factor in their findings (for review see Mendrek 2015), possibly jeopardizing their results’ interpretation.

It has been shown that a gamble is judged to be more attractive when its zero outcome is designated as “losing $0” rather than “winning $0,” an instance of what we refer to as the mutable-zero effect

The framing of nothing and the psychology of choice. Marc Scholten, Daniel Read, Neil Stewart. Journal of Risk and Uncertainty, December 3 2019. https://link.springer.com/article/10.1007/s11166-019-09313-5

Abstract: Zero outcomes are inconsequential in most models of choice. However, when disclosing zero outcomes they must be designated. It has been shown that a gamble is judged to be more attractive when its zero outcome is designated as “losing $0” rather than “winning $0,” an instance of what we refer to as the mutable-zero effect. Drawing on norm theory, we argue that “losing $0” or “paying $0” evokes counterfactual losses, with which the zero outcome compares favorably (a good zero), and thus acquires positive value, whereas “winning $0” or “receiving $0” evokes counterfactual gains, with which the zero outcome compares unfavorably (a bad zero), and thus acquires negative value. Moreover, we propose that the acquired value of zero outcomes operates just as the intrinsic value of nonzero outcomes in the course of decision making. We derive testable implications from prospect theory for mutable-zero effects in risky choice, and from the double-entry mental accounting model for mutable-zero effects in intertemporal choice. The testable implications are consistently confirmed. We conclude that prevalent theories of choice can explain how decisions are influenced by mutable zeroes, on the shared understanding that nothing can have value, just like everything else.

Keywords: Descriptive invariance Norm theory Counterfactuals Zero outcomes Risk and time Prospect theory Double-entry mental accounting model
JEL Classifications: D00 D90 D91



4 General discussion
The valence of a zero event depends on its “irrelevant” description: It “feels better” to
lose or pay nothing than to win or receive nothing. A negative wording (lose, pay) sets
up a norm of negative events, with which the zero event compares favorably, while a
positive wording (win, receive) sets up a norm of positive events, with which the zero
event compares unfavorably, so that a negative wording acquires a more positive tone
than a positive wording. Descriptive invariance requires from us that this should not
affect our decisions, but we have shown that it does, among a fair number of us at least.
To others among us, the framing of zero events may actually be irrelevant. The
mutable-zero effect is indeed small; yet, it is a reliable phenomenon. And if one thinks
of the alternative descriptions of a zero outcome as a minimal manipulation, the small
effect may actually be considered quite impressive (Prentice and Miller 1992).
Descriptive invariance, along with dominance, is an essential condition of rational
choice (Tversky and Kahneman 1986), and it has seen a number of violations,
commonly referred to as framing effects. A stylized example of framing is the adage
that optimists see a glass of wine as half full, while pessimists see it as half empty. And
if the wine glass is half full, and therefore half empty, then these are complementary
descriptions of the same state of the world, so that, normatively, using one or the other
should not matter for judgment and choice (Mandel 2001)—but it does.

4.1 Counterfactuals versus expectations
Life often confronts us with zero events. A bookstore may offer us “free shipping.” Our
employermay grant us “no bonus.”We are pleased to pay $0 to the bookstore, and this may be
because we expected to pay something but did not.We are not pleased to receive $0 from our
employer, and this may be because we expected to receive something but did not (Rick and
Loewenstein 2008). In norm theory, Kahneman and Miller (1986) suggested that reasoning
may not only flow forward, “from anticipation and hypothesis to confirmation or revision,” but
also backward, “from the experience to what it reminds us of or makes us think about” (p.
137). In the latter case, “objects or events generate their own norms by retrieval of similar
experiences stored in memory or by construction of counterfactual alternatives” (p. 136). Thus,
“free shipping” may sound pleasant because it reminds us of occasions on which we were
charged shipping fees, and “no bonus” may sound unpleasant because it reminds us of
occasions on which we were granted bonuses; not so much because we expected to pay or
receive something. Of course, both norms and expectations may influence our feelings, and
may be difficult to disentangle in many real-life situations. Kahneman and Miller’s (1986)
intention with norm theory was “not to deny the existence of anticipation and expectation but
to encourage the consideration of alternative accounts for some of the observations that are
routinely explained in terms of forward processing” (p. 137, emphasis added). Our intention
was to compile a set of observations that cannot reasonably be explained in terms of forward
processing, which therefore constitute the clearest exposure of norms.

4.2 Expectations in decision theory
We have incorporated counterfactuals into theories of choice, so as to predict the effects
of mutable zeroes when people face risk and when people face time. Traditionally,
decision theory has ignored counterfactuals, but expectations play a role in most
theories of decision under risk. While prospect theory sacrifices the expectation
principle from EU, by assigning a decision weight w(p) to probability p of an outcome
occurring, other formulations have maintained the expectation principle but modified
the utility function. For instance, the utility function has been expanded with anticipated
regret and rejoicing as they result from comparisons between the possible outcomes
of a gamble and those that would occur if one were to choose differently (Bell 1982,
1983; Loomes and Sugden 1982). Similarly, the utility function has been expanded
with anticipated emotions as they result from comparisons between the possible
outcomes of a gamble with the expected value of the gamble: Anticipated disappointment
when “it could come out better,” and anticipated elation when “it could come out
worse” (Bell 1985; Loomes and Sugden 1986). Zero outcomes acquire value in the
same way as nonzero outcomes do: Either from between-gamble or within-gamble
comparisons. Thus, a zero outcome acquires negative value (by regret or disappointment)
if the comparison is with a gain, and positive value (by rejoicing or elation) if the
comparison is with a loss. In our analysis, however, zero outcomes are unique, in that
only they elicit counterfactual gains and losses, which will then serve as a reference
point for evaluating the zero outcomes themselves. Nonetheless, in Experiment 3,
dealing with zero outcomes in intertemporal choice, we obtained a result suggesting
that between-prospect comparisons of zero and nonzero outcomes also affected choice.

4.3 The framing of something, the framing of nothing
The investigation of framing effects in judgment and decision making began with
Tversky and Kahneman’s (1981) Asian Disease Problem, in which the lives of 600
people are threatened, and life-saving programs are examined. One group of participants
preferred a program that would save 200 people for sure over a program that
would save 600 people with a 1/3 probability, but save no people with a 2/3 probability.
Another group of participants preferred a program that would let nobody die with a
probability of 1/3, but let 600 people die with a 2/3 probability, over a program that
would let 400 people die for sure. Prospect theory ascribes this result to reference
dependence, i.e., v(0) = 0, and diminishing sensitivity, i.e., v is concave over gains, so
that v(600) < 3v(200), which works against the gamble, and convex over losses, so that
v(−600) > 3v(−200), which works in favor of the gamble.
Our interpretation is that some of the action may lie in the zero outcomes, rather than
the nonzero outcomes. Specifically, “save no people” brings to mind saving some
people, with which saving no people compares unfavorably, thus working against the
gamble. Similarly, “let nobody die” brings to mind letting somebody die, with which
letting nobody die compares favorably, thus working in favor of the gamble. Reference
dependence is fine, but designating zero outcomes means that v(0) ≠ 0, because the
reference point is no longer the status quo, but rather something imagined.
There is no shortage of competing views on framing effects (for one of many
discussions, see Mandel 2014), and our norm-theory approach to the Asian Disease
Problem is a partial explanation at best. Indeed, the reversal from an ample majority
(72%) choosing the safe option in the positive frame (saving lives) to an ample majority
(78%) choosing the risky option in the negative frame (giving up lives) is a large effect,
whereas the mutable-zero effect is a small effect, unlikely to be the sole responsible for
Tversky and Kahneman’s (1981) result. However, judgments and decisions are influenced
by the framing of zero outcomes, and we have shown that prevalent theories of
choice, Kahneman and Tversky’s (1979) prospect theory and Prelec and Loewenstein’s
(1998) double-entry mental accounting model, can explain how decisions are influenced
by mutable zeroes, on the shared understanding that nothing can have value, just
like everything else.

Results provide a cautionary tale for the naïve application of VAMs to teacher evaluation and other settings; they point to the possibility of the misidentification of sizable teacher “effects”where none exist

Teacher Effects on Student Achievement and Height: A Cautionary Tale. Marianne Bitler, Sean Corcoran, Thurston Domina, Emily Penner. NBER Working Paper No. 26480, November 2019. https://www.nber.org/papers/w26480

Abstract: Estimates of teacher “value-added” suggest teachers vary substantially in their ability to promote student learning. Prompted by this finding, many states and school districts have adopted value-added measures as indicators of teacher job performance. In this paper, we conduct a new test of the validity of value-added models. Using administrative student data from New York City, we apply commonly estimated value-added models to an outcome teachers cannot plausibly affect: student height. We find the standard deviation of teacher effects on height is nearly as large as that for math and reading achievement, raising obvious questions about validity. Subsequent analysis finds these “effects” are largely spurious variation (noise), rather than bias resulting from sorting on unobserved factors related to achievement. Given the difficulty of differentiating signal from noise in real-world teacher effect estimates, this paper serves as a cautionary tale for their use in practice.

6   Discussion
Schools and districts across the country want to employ teachers who can best help students to learn, grow, and achieve academic success. Identifying such individuals is integral to schools' successbutis also difficult to do in practice. In the face of data and measurement limitations, school leaders and state education departments seek low-cost, unbiased ways to observe and monitor the impact that their teachers have on students. Although many have criticized the use of VAMs to evaluate teachers, they remain a widely-used measure of teacher performance. In part, their popularity is due to convenience-while observational protocols which send observers to every teacher's classroom require expensive training and considerable resources to implement at scale, VAMs use existing data and can be calculated centrally at low cost. Further, VAMs are arguably less biased than many other evaluation methods that districts might use instead (Bacher-Hicks et al. 2017; Harris et al. 2014; Hill et al. 2011).

Yet questions remain about the reliability, validity, and practical use of VAMs. This paper interrogates concerns raised by prior research on VAMs and raises new concerns about the use of VAMs in career and compensation decisions. We explore the bias and reliability of commonlyestimated VAMs by comparing estimates of teacher value-added in mathematics and ELA with parallel estimates of teacher value-added on a well-measured biomarker that teachers should not impact: student height. Using administrative data from New York City, we find estimated teacher “effects”on height that are comparable in magnitude to actual teacher effects on math and ELA achievement, 0.22:compared to 0.29:and0.26:respectively. On its face, such results raise concerns about the validity of these models.

Fortunately, subsequent analysis finds that teacher effects on height are primarily noise, rather than bias due to sorting on unobserved factors. To ameliorate the effect of sampling error on value-added estimates, analysts sometimes “shrink” VAMs, scaling them by their estimated signal-to-noise ratio. When we apply the shrinkage method across multiple years of data from Kane and Staiger (2008), the persistent teacher “effect”on height goes away, becoming the expected (and known) mean of zero. This procedure is not always done in practice, however, and requires multiple years of classroom data for the same teachers to implement. Of course, for making hiring and firing decisions, it seems important to consider that value added measures which require multiple years of data to implement will likely permit identification of persistently bad teachers, but not provide a performance evaluation metric that can be met by teachers trying to improve their performance. In more realistic settings where the persistent effect is not zero, it is less clear that shrinkage would have a major influence on performance decisions, since it has modest effects on the relative rankings of teachers.

Taken together, our results provide a cautionary tale for the naïve application of VAMs to teacher evaluation and other settings. They point to the possibility of the misidentification of sizable teacher “effects”where none exist. These effects may be due in part to spurious variation driven by the typically small samples of children used to estimate a teacher's individual effect.