# How to make surveys cheaper and better

Let’s say that a survey company would like to make a poll about the upcoming American election. How do they go about it? The goal for them would be to make the survey both cheaper and better. To make the survey more cost efficient, it is important to know if it is possible to reduce the sample size in the survey without increasing the errors. Edgar Bueno, doctoral student, researches a method to do that.

# How to solve problems with non-response

As the non-response rates of important studies continue to rise, the quality of the results are getting worse. Per Gösta Andersson from the Department of Statistics researches together with Carl-Erik Särndal, Professor Emeritus of Statistics Sweden, into a method that mitigates the effect of non-response.

# A better way of predicting the economy

For planners, it is very important to know how fast the economy is growing in order to make the right decisions. Mahmood Ul Hassan, PhD scholar in statistics, has recently published a paper on how to estimate economic growth in a better way.

# Statistics can help doctors make the right decisions

What is the proportion of patients dying from a certain disease? Or, would a certain treatment increase a patient’s survival of that disease? Parfait Munezero, doctoral student, researches into the area in statistics that can answer these types of questions.

# How many times should we try to call someone?

To try to make contact with people, alternative to classic random sampling and extrapolation - this is our ongoing research on sampling in the profile area Official statistics.

# This is how the new Big Data research works

A workhorse within the Bayesian inference world of statistics. This is how Matias Quiroz, doctor of statistics, describes the MCMC algorithm that is used to estimate complex models. He has shown how this it can be speeded up - an advantage when working with Big Data.

# Not true that more siblings cause lower intelligence

The more siblings there are in a family, the lower the cognitive ability of the children. This has been demonstrated in several studies. But research by Linda Wänström and Gebrenegus Ghilagaber shows that the relationship is not that simple.

# Statistics can improve the treatment of rare diseases

Those worst affected by rare diseases are children. Frank Miller, Associate Professor of statistics at Stockholm University researches in two different projects that aim to improve the chances these patients have for receiving treatment.

# Statistical Diagnostics in Multilevel Models

**Tatjana von Rosen** has grants from Swedish Research Council.

Multilevel modelling is a widely accepted tool for analyzing data which are organized into hierarchies or other complex classification structures. The overall purpose of the project is to extend existing statistical diagnostic tools and develop new ones for in multilevel models and provide data analysts with routine methods to assess the influence of observations on the model parameters and model fit. The project has the following specific aims: (i) to define single and multiple outliers and influential observations in multilevel models with respect to the dependent variable and covariates, respectively. (ii) to develop statistical diagnostics to detect outliers and influential observations defined in (i). (iii) To study properties of the statistical diagnostics obtained in (ii).