Master Thesis

We have now finished our master thesis and handed it in for evaluation on 1 February 2007. The title of the thesis is:

Diabetikeres præferencer for udstyr til administration af insulin: Et discrete choice experiment

The paper is written in Danish, but with an English summary.

The English title is:

Diabetic patients’ preferences for insulin administration systems: A discrete choice experiment

A date has not yet been set for the oral defence of the thesis.

Below you will find a brief description of the thesis as well as a list of downloads. Even though the paper is written in Danish, it should be possible for all to interpret the results given in the tables. Also, documentation of the statistical programming has been kept entirely in English.

About the thesis

We have used the seminar as a literature study and a technical study to prepare ourselves for the thesis. The thesis, on the other hand, is an empirical study of alternative ways to administer insulin. The study is carried out using a so-called discrete choice experiment (DCE, the method also described in the seminar paper). Our focus has been on diabetics’ willingness-to-pay for various properties of the device used to administer insulin – i.e. insulin pen (syringe) and insulin inhaler.

The empirical study was financed by Novo Nordisk A/S. This means that Novo owns the rights to the collected data, and, hence, that data is not ours to give away. Instead, we provide you with a fictive data set, which you can use to verify the appropriateness of the statistical programming. Obviously, you will not be able to reproduce the actual results. Results are documented in appendices and in output from Stata, both of which can be
downloaded below.

You can read the full English summary here.

Download paper, appendices and documentation

Below you will find links to the paper itself (in pdf format), programming code used to carry out the analysis and output from the analysis (we are using Stata 9, but code and output are ordinary ASCII text, so you should be able to read it with any text editor). The paper itself is written in Danish, but with an English summary. Below you will also find all three versions of the questionnaire as well as a couple of other documents not included in the paper (or in appendices).

As mentioned, we are not able to supply you with the data set used in the analysis, since we do not own the rights to data. Instead we provide you with a fictive data set.

You can download the following files:

We will consider adding a number of notes, describing specific parts of the process of planning, designing and carrying out the empirical study, to the above list of downloads; but only if these are requested.

Alternative method, if you have Stata installed on your computer:

If you have the statistical software package Stata installed on your computer, you can obtain the paper as well as programming documentation in a much easier manner (equivalent to the paper and the statistical analysis zip archive listed above). Just write the following three lines of code in Stata’s Command Prompt (one at a time):

. net from
. net describe speciale
. net get speciale

Please write each line exactly as it is spelled out above (except, without the prefixed period. That’s just Stata convention for showing commands in output). All files will be downloaded to Stata’s current working directory, so please be sure to change the directory, if you wish to download the files to another location on your computer’s hard drive (-pwd- will show you Stata’s current working directory, which you can then change using -cd-).


– Majbritt & Jesper

To the top of the page

Comments are closed.