Skip to content

Post-stratified longitudinal item response model for trust in state institutions in Europe

This is a guest post by Marta Kołczyńska:

Paul, Lauren, Aki, and I (Marta) wrote a preprint where we estimate trends in political trust in European countries between 1989 and 2019 based on cross-national survey data.

This paper started from the following question: How to estimate country-year levels of political trust with data from surveys that (a) mostly have the same trust questions but measured with ordinal rating scales of different lengths, and (b) mostly have samples that aim to be representative for general adult populations, but this representativeness is likely reached to different degrees?

Our solution combines:

  1. item response models of responses to trust items that account for the varying scale lengths across survey projects,
  2. splines to model changes over time,
  3. post-stratification by age, sex, and education.

In the paper we try to explain all the modeling decisions, so that the paper may serve as a guide for people who want to apply similar methods or — even better — extend and improve them.

We apply this approach to data from 12 cross-national projects (1663 national surveys) carried out in 27 European countries between 1989 and 2019. We find that (a) political trust is pretty volatile, (b) there has not been any clear downward trend in political trust in Europe in the last 30 years, although trust did decline in many Central-East European countries in the 1990s, and there was a visible dip following the 2008 crisis in countries that were hit most, followed by at least partial recovery. Below are estimated levels of political trust for the 27 countries (see the preprint for more details on differences in political trust by sex, age, and education):

Estimated political trust in Europe

The modeling was done in brms thanks to some special features that Paul wrote, and overall this is one of the projects that would not have been possible without Stan.

One of the main obstacles we faced was the limited availability of population data for post-stratification. In the end we used crude education categories (less than high school, high school or above – also because of the incoherent coding of education in surveys), combined Eurostat data with harmonized census samples from IPUMS International, and imputed values for the missing years.

We think our approach or some of its components can be more broadly applied to modeling attitudes in a way that addresses issues of measurement and sample representativeness.


  1. Ebe Ouattara says:

    Terrific work! And a great example. Looking forward to learning from it and applying it to the consequences of trust.

  2. Andrew says:


    Some quick comments on the graphs:

    1. Maybe mix up the colors a bit. It’s hard for me to untangle the very similarly colored curves in each plot. You could also do 6 plots instead of just 4 so that there are fewer curves per plot.

    2. Arrange the 4 plots in a logical way, not alphabetically! I suggest western Europe on the upper left, southern Europe on the lower left, northern Europe on the upper right, and central and eastern Europe in the lower right.

    3. You’ll get a bit more real estate if you just have the y-axis go up to 1.5. Right now it goes up to 2.2. If zero is important and your concern is that you want balance relative to zero, you can get that by putting a solid horizontal line at zero in each plot.

    • jim says:

      Good comments,

      I agree w/ dividing the “central and eastern” into two plots and the “western” into two plots, but I like the general scheme.

      But as long as we’re exploring possibilities, another possibility for the layout is a map background showing each of the (revised to) six blocks of countries, and showing the plots as sideways “T’s” like this ├─ with callouts to the country blocks, rather than squares. Axes with tics only, then a reference chart in one corner showing the values of the tics. I think this works because the numbers are pretty simple, just decades and 0-1-2.

  3. Josh says:

    I like how the manuscript used probit for the graded response model rather than logit. It’s a minor point but it’s cleaner mathematically – in particular, being closed under Gaussian convolution. Unfortunately, due to historical reasons, the logit is in much wider use for this type of model. I had thought to switch to using probit but ultimately decided against it because I lack bravery.

Leave a Reply