Lars P. Syll

.[embedded content]

Language — an amazing thing .[embedded content]

Interpretation of regression results When econometric and statistical textbooks present simple (and multiple) regression analysis for cross-sectional data, they often do it with regressions like “regress test score (y) on study hours (x)” and get the result y = constant + slope coefficient*x + error term. When speaking of increases or decreases in x in these interpretations, we have to remember that it is a question of cross-sectional data and ‘increases’...

Causal mediation and the limits of linear models (student stuff) .[embedded content]

.[embedded content] Spending a couple of hours going through a JEL survey of modern research on the gender wage gap, yours truly was struck almost immediately by how little that research really has accomplished in terms of explaining gender wage discrimination. With all the heavy regression and econometric alchemy used, wage discrimination is somehow more or less conjured away … Trying to reduce the risk of having established only ‘spurious relations’ when dealing with...

Tant Bergström hälsar och tackar från Caymanöarna … .[embedded content]

How to get from association to causation — Diversity of Evidence vs RCTs .[embedded content]

.[embedded content] (In case you’re not familiar with the ongoing debate on how to analyse mediation and moderation (interaction) in causal models, yours truly suggests you start by reading Judea Pearl’s The Book of Why (2018), chapter 9.)

. I Sverige låter vi år 2021 friskolekoncerner med undermålig verksamhet få plocka ut skyhöga vinster — vinster som den svenska staten gladeligen låter dessa koncerner ta av vår skattefinansierade skolpeng. Ett flertal undersökningar har på senare år  visat att det system vi har i Sverige med vinstdrivande skolor leder till att våra skolor blir allt mindre likvärdiga — och att detta i sin tur bidrar till allt sämre resultat. Ska vi råda bot på detta måste vi ha ett skolsystem...

On mediation and causality “Mediation analysis” is this thing where you have a treatment and an outcome and you’re trying to model how the treatment works: how much does it directly affect the outcome, and how much is the effect “mediated” through intermediate variables … In the real world, it’s my impression that almost all the mediation analyses that people actually fit in the social and medical sciences are misguided: lots of examples where the...