One-day workshop: Survey methodology for sensitive questions using
randomized response

Tutors: Prof. Dr. Peter G.M. van der Heijden (Utrecht University and
Lancaster University) and 
Dr. Maarten Cruyff (Utrecht University)

Thursday 22 April 2010, 10-5pm

Venue: Postgraduate Statistics Centre, Lancaster University

To book a place please go to: http://shortcourses.maths.lancs.ac.uk/

=================================================================
It is well-known that, in surveys, questions about sensitive behaviour
may not yield truthful responses. This master class will discuss recent
developments in the design of randomized response studies and the
analysis of randomized response data. For the design a best practice is
shown that has been regularly used in the Netherlands, mainly in the
context of social benefit fraud. 

Many statistical tools are currently available for the analysis of
randomized response data, such as logistic regression (useful when the
sensitive question is the response variable to be related to explanatory
variables), loglinear analysis (useful to investigate relations between
various sensitive questions), item response theory models (useful for
investigating the extent of individual sensitive behaviour from a number
of sensitive questions) and count data models (counting the number of
sensitive behaviours). In addition, extensions to the above statistical
models are proposed that accommodate respondents that do not follow the
randomized response design by saying "no" to whatever sensitive question
is asked (the untruthful-responses-problem). 

THE INSTRUCTORS
Peter van der Heijden and Maarten Cruyff have published several papers
on the development of the design of randomized response studies and the
analysis of randomized response data, and published in journals such as
JRSS Series A and Series B, Sociological Methods and Research,
Psychometrika, the Annals of Applied Statistics, Applied Statistics and
Journal of Applied Econometrics. They have carried out a series of
surveys for the Dutch ministry of employment into benefit fraud. Their
website is found at www.randomisedresponse.nl. 

FURTHER DETAILS
In surveys, questions about sensitive behaviour may not yield truthful
responses. To solve this problem, Warner (1965) proposed the randomized
response method as a survey tool to get more honest answers to sensitive
questions (Warner, 1965). In the original randomized response approach,
respondents are provided with two statements, A and B, with statement A
being  the complement of statement B. For example, statement A is 'I
used hard drugs last year' and statement B is 'I did not use hard drugs
last year'. A randomizing device, for instance, in the form of a pair of
dice determines whether statement A or B is to be answered. The
interviewer records the answer `yes' or `no' without knowing the outcome
of the randomizing response device. Thus the interviewee's privacy is
protected but it is still possible to calculate the probability that the
sensitive question (A and not-B) is answered positively.

Recent meta-analyses (Lensvelt et al., 2005, SMR) show that randomized
response methods significantly outperform more direct ways of asking
sensitive questions. Importantly, the validity increases with the
sensitivity of the topic under investigation.

The instructors argue that in the past it has not been used as often as
it could be, and that there are probably three reasons for this.
Firstly, randomized response research is expensive as larger sample
sizes are needed to obtain the same precision of estimates. Secondly,
researchers mistakenly believe that randomized response only allows for
the estimation of the prevalence of sensitive behaviour. Thirdly,
randomized response is not believed to be solving the untruthful-answer
problem. These three reasons can be rebutted as follows: firstly, data
collection over the internet has become easier over the years and can
solve the sample size problem. Secondly, as is indicated above, a whole
range of tools for the analysis of randomized response data exists. And
thirdly, new statistical methodology is available to handle respondents
that do not comply to the randomized response design.

Time schedule:
9.30 - 10.00 Registration
10.00 - 12.00 A best practice for randomized response research (with
short breaks)
13.00 - 16.00 Statistical models for the analysis of randomized response
data (with short breaks)
16.00 - 17.00  Discussion and conclusion

For any enquiries please contact: Deborah Stewart, PSC secretary,
[log in to unmask]

You may leave the list at any time by sending the command

SIGNOFF allstat

to [log in to unmask], leaving the subject line blank.