UTS site search

Professor Deborah Street

Professor of Statistics, School of Mathematical Sciences
Associate Member, Centre for Health Economics Research & Evaluation
BSc (Hons) (UQ), PhD (Syd)
+61 2 9514 2251

Research Interests

The general area of my research is in the construction of designed experiments.

Most of my research has been motivated by design questions that arise from practical problems. I have developed optimal designs for repeated measurements (changeover) designs, incomplete factorial designs in which all main effects can be estimated and I have designed designs for training dispute resolution councillors. My most recent work has been the construction of optimal and small (near-)optimal discrete choice experiments (DCEs).

DCEs are used extensively in marketing, transportation research, environmental economics and health economics. Dr Leonie Burgess and I have developed constructions for optimal designs for the multinomial logit (MNL) model for estimation of main effects for situations in which the attributes may have any number of levels and the choice sets may be of any size. We have derived the determinant of the information matrix for the estimation of main effects and two-factor interactions and we have tables of optimal designs for some combinations of attribute levels for various choice set sizes. We have developed a heuristic which produces small near-optimal designs when the optimal designs are unmanageably large. Our research monograph with these results was published by Wiley.

Can supervise: Yes

Book Chapters

Louviere, J.J., Viney, R.C., Street, D. & Burgess, L.B. 2012, 'Design of Choice Experiments in Health Economics' in Andrew M. Jones (ed), The Elgar Companion to HEalth Econimics, Second Edition, Edward Elgar Publishing Limited, Northampton, MA, USA, pp. 447-462.
View/Download from: OPUS
In many areas of applied economics, economists use observations of actual choices, or revealed preference (RP) data, to model behaviour. Individuals are assumed to make utility maximizing choices, and utility functions are estimated by analysing observed choices. A need for stated preference (SP) data arises when there are limited or no RP data available, because the good or service is new, not provided in a market context, or there is insufficient variability in choice attributes to obtain reliable estimates of their effects. Discrete choice experiments (DCEs) are an SP method of interest in health economics because they allow analysis of preferences for complex, multi-attribute goods like health care.
Street, D. & Burgess, L.B. 2012, 'Designs for choice experiments for the multinomial logit model' in Klaus Hinkelmann (ed), Design and Analysis of Experiments: Special Designs and Applications Volume 3, John Wiley & Sons, United States, pp. 331-378.
View/Download from: OPUS |
People make choices all the time; some of these are minor, like deciding what clothes to wear to work today, but some are major and of interest to governments and businesses. Governments might be interested in modelling demand for health services in the future, for instance, or in assessing the likely impact on the electorate of a decision to allow mining or logging in national parks. Businesses want to predict the likely market for new goods and services. To get information about products or services that do not yet exist, an experimental approach is appropriate. Such experiments are called +stated preference+ or +stated choice+ experiments. This chapter provides an overview of the best way to design generic stated preference choice experiments, from a mathematical perspective. A more extensive discussion appears in Street and Burgess (2007).
Street, D., Burgess, L.B., Viney, R.C. & Louviere, J.J. 2008, 'Designing Discrete Choice Experiments for Health Care' in Ryan,M; Gerard, K; Amaya-Amaya, M. (eds), Using Discrete Choice Experiments to Value Health and Health Care, Springer, Dordrecht, The Netherlands, pp. 47-72.
View/Download from: OPUS | Publisher's site
As noted in Chapter 1, the application of discrete choice experiments (DCEs) in health economics has seen an increase over the last few years.While the number of studies using DCEs is growing, there has been relatively limited consideration of experimental design theory and methods. Details of the development of the designed experiment are rarely discussed. Many studies have used small fractional factorial designs (FFDs), generated with commercial design software packages, e.g. orthogonal main effects plans (OMEPs), sometimes manipulated in ad hoc ways (e.g. randomly pairing up scenarios or taking one scenario from the design and combining it with every other scenario). Such approaches can result in designs with unknown statistical design properties, in particular with unknown correlations between parameter estimates.
Burgess, L.B., Street, D., Viney, R.C. & Louviere, J.J. 2006, 'Design of choice experiments in health economics' in Jones, A M (eds), The Elgar Companion to Health Economics, Edward Elgar, Cheltenham, UK, pp. 415-426.
View/Download from: OPUS
Louviere, J.J., Street, D. & Burgess, L.B. 2004, 'A 20+ years' retrospective on choice experiments' in Wind, Y; Green, P E (eds), Marketing Research and Modeling: Progress and Prospects, Kluwer Academic Publishers, Boston, pp. 201-214.
Paul Green inspired many of us who work in conjoint analysis and related areas, such as stated preference discrete choice experiments, and all of us who undertake research in understanding and modelling preferences have benefited from his work. Knowing Paul Green, his impact and contributions will continue during his "returement". Our thanks to him for so many of the advances that we now often take for granted, but without which we would not be where we are today. In tribute to Paul Green, this paper provides a bried history of choice experiments, how the results of choice experiments relate to real market behaviour,. and progress that has been made int he design of choice experiments. We concludee by discussing several aspects of choice experimens, such as the interacctions of subjects with experiments, model complexity and the scientific understanding of choice behaviour compared with mere predictions of choices.
Cheng, Y.W., Street, D. & Wilson, W.H. 2003, 'Two-stage generalised simulated annealing for the construction of change-over designs' in Wallis, WD (eds), Designs 2002. Further Computational and Constructive Design Theory, Kluwer Academic Publishers, Dordrecht, Netherlands, pp. 69-79.
Louviere, J.J. & Street, D. 2000, 'Stated-preference Methods' in David A Hensher & Kenneth J Button (eds), Handbook of Transport Modelling, Pergamon Press, Amsterdam, Netherlands, pp. 131-143.


Street, D. & Burgess, L.B. 2007, The Construction of Optimal Stated Choice Experiments: Theory and Methods, 1, John Wiley and Sons, Hoboken, New Jersey.
View/Download from: OPUS

Conference Papers

Tomkin, D.F., Thomas, L.E., Day, M.B., Burke, P.F., Franklin, J.B., Smith, G.B., Louviere, J.J. & Street, D. 2007, 'Solar Light for rooms without windows', Sustainable Innovation 07, Farnham, Surrey, UK, October 2007.
Louviere, J.J., Burke, P.F., Street, D., Burgess, L.B. & Marley, A.A. 2006, 'Dicrete choice surveys: Improving completion rates and getting better data', Asia-Pacific Quantitative Methods in Marketing Conference, Sydney, Australia, October 2006.

Journal Articles

Viney, R.C., Norman, R., Brazier, J.E., Cronin, P.A., King, M.T., Ratcliffe, J. & Street, D. 2014, 'An Australian discrete choice experiment to value EQ-5D health states', Health Economics, vol. 23, no. 6, pp. 729-742.
Conventionally, generic quality-of-life health states, defined within multi-attribute utility instruments, have been valued using a Standard Gamble or a Time Trade-Off. Both are grounded in expected utility theory but impose strong assumptions about the form of the utility function. Preference elicitation tasks for both are complicated, limiting the number of health states that each respondent can value and, therefore, that can be valued overall. The usual approach has been to value a set of the possible health states and impute values for the remainder. Discrete Choice Experiments (DCEs) offer an attractive alternative, allowing investigation of more flexible specifications of the utility function and greater coverage of the response surface. We designed a DCE to obtain values for EQ-5D health states and implemented it in an Australia-representative online panel (n?=?1,031). A range of specifications investigating non-linear preferences with respect to time and interactions between EQ-5D levels were estimated using a random-effects probit model. The results provide empirical support for a flexible utility function, including at least some two-factor interactions. We then constructed a preference index such that full health and death were valued at 1 and 0, respectively, to provide a DCE-based algorithm for Australian cost-utility analyses.
Hall, J.P., Kenny, P.M., Hossain, I., Street, D. & Knox, S.A. 2014, 'Providing informal care in terminal illness: an analysis of preferences for support using a discrete choice experiment', Medical Decision Making, vol. 34, no. 6, pp. 731-745.
Background: The trend for terminally ill patients to receive much of their end-of-life care at home necessitates the design of services to facilitate this. Care at home also requires that informal care be provided by family members and friends. This study investigated informal carers' preferences for support services to aid the development of end-of-life health care services. METHODS: This cross-sectional study used 2 discrete choice experiments to ascertain the preferences of carers supporting patients with different levels of care need, determined by the assistance needed with personal care and labeled High Care (HC) and Low Care (LC). The sample included 168 informal carers of people receiving palliative care at home from 2 palliative care services in Sydney, Australia. Data were collected in face-to-face interviews; carers chose between 2 hypothetical plans of support services and their current services. Data were analyzed with generalized multinomial logit models that were used to calculate the impact of each attribute on the probability of a carer choosing a service plan. RESULTS: Preferred support included nursing services; the probability of choosing a plan increased significantly if it included nurse home visits and phone advice (P < 0.001). HC carers also wanted doctor home visits, home respite, and help with personal care (P < 0.05), and LC carers wanted help with household tasks, transport, and a case coordinator (P < 0.001). On average, both groups of carers preferred their current services, but this varied with characteristics of the carer and the caregiving situation. CONCLUSIONS: The most valued services are those that support carers in their caregiving role; however, supportive care preferences vary with the different circumstances of patients and carers.
Norman, R., Viney, R.C., Brazier, J.E., Burgess, L.B., Cronin, P.A., King, M.T., Ratcliffe, J. & Street, D. 2014, 'Valuing SF-6D health states using a Discrete Choice Experiment', Medical Decision Making, vol. 34, no. 6, pp. 773-786.
Background. SF-6D utility weights are conventionally produced using a standard gamble (SG). SG-derived weights consistently demonstrate a floor effect not observed with other elicitation techniques. Recent advances in discrete choice methods have allowed estimation of utility weights. The objective was to produce Australian utility weights for the SF-6D and to explore the application of discrete choice experiment (DCE) methods in this context. We hypothesized that weights derived using this method would reflect the largely monotonic construction of the SF-6D. Methods. We designed an online DCE and administered it to an Australia-representative online panel (n = 1017). A range of specifications investigating nonlinear preferences with respect to additional life expectancy were estimated using a random-effects probit model. The preferred model was then used to estimate a preference index such that full health and death were valued at 1 and 0, respectively, to provide an algorithm for Australian cost-utility analyses. Results. Physical functioning, pain, mental health, and vitality were the largest drivers of utility weights. Combining levels to remove illogical orderings did not lead to a poorer model fit. Relative to international SG-derived weights, the range of utility weights was larger with 5% of health states valued below zero. Conclusions. DCEs can be used to investigate preferences for health profiles and to estimate utility weights for multi-attribute utility instruments. Australian cost-utility analyses can now use domestic SF-6D weights. The comparability of DCE results to those using other elicitation methods for estimating utility weights for quality-adjusted life-year calculations should be further investigated.
Doiron, D., Hall, J.P., Kenny, P.M. & Street, D. 2014, 'Job preferences of students and new graduates in nursing', Applied Economics, vol. 46, no. 9, pp. 924-939.
View/Download from: Publisher's site
This article investigates the preferences of student and newly graduated nurses for pecuniary and nonpecuniary aspects of nursing jobs. It is the first study applying methods based on discrete choice experiments to a developed country nursing workforce. It is also the first to focus on the transition through university training and into work. This is particularly important as junior nurses have the lowest retention levels in the profession. We sample 526 individuals from nursing programmes in two Australian universities. Flexible and newly developed models combining heteroscedasticity with unobserved heterogeneity in scale and preference weights are estimated.Overall, salary remains the most important feature in increasing the probability that a job will be selected. `Supportive management/staff+ and `quality of care+ follow as the most important attributes from a list of 11 nonpecuniary characteristics. However, the subset of new graduates rank `supportive management/staff+ above salary increases, emphasizing the importance of a supportive workplace in the transition from university to the workplace. We find substantial preference heterogeneity and some attributes, such as the opportunity for clinical rotations, are found to be attractive to some nurses while seen as negative by others. Nursing retention could be improved by designing different employment packages to appeal to these different tastes.
Norman, R., Hall, J.P., Street, D. & Viney, R.C. 2013, 'Efficiency and equity: A stated preference approach', Health Economics, vol. 22, no. 5, pp. 568-581.
View/Download from: OPUS | Publisher's site
Outcome measurement in the economic evaluation of health care considers outcomes independent of to whom they accrue. This article reports on a discrete choice experiment designed to elicit population preferences regarding the allocation of health gain between hypothetical groups of potential patients. A random-effects probit model is estimated, and a technique for converting these results into equity weights for use in economic evaluation is adopted. On average, the modelling predicts a relatively high social value on health gains accruing to nonsmokers, carers, those with a low income and those with an expected age of death less than 45?years. Respondents tend to favour individuals with similar characteristics to themselves. These results challenge the conventional practice of assuming constant equity weighting. For decision makers, whether a formal equity weighting system represents an improvement on more informal approaches to weighing up equity and efficiency concerns remains uncertain
Knox, S.A., Viney, R.C., Gu, Y., Hole, A., Fiebig, D.G., Street, D., Haas, M.R., Weisberg, E. & Bateson, D. 2013, 'The effect of adverse information and positive promotion on women's preferences for prescribed contraceptive products', Social Science & Medicine, vol. 83, pp. 70-80.
View/Download from: OPUS | Publisher's site
Recent rapid growth in the range of contraceptive products has given women more choice, but also adds complexity to the resultant decision of which product to choose. This paper uses a discrete choice experiment (DCE) to investigate the effect of adverse information and positive promotion on women's stated preferences for prescribed contraceptive products. In November 2007, 527 Australian women aged 18-49 years were recruited from an online panel. Each was randomly allocated to one of three information conditions. The control group only received basic information on contraceptive products. One treatment group also received adverse information on the risks of the combined oral pill. The other group received basic information and promotional material on the vaginal ring, newly introduced into Australia and on the transdermal patch, which is unavailable in Australia. Respondents completed 32 choice sets with 3 product options where each option was described by a product label: either combined pill, minipill, injection, implant, hormonal IUD, hormonal vaginal ring, hormonal transdermal patch or copper IUD; and by the attributes: effect on acne, effect on weight, frequency of administration, contraceptive effectiveness, doctor's recommendation, effect on periods and cost. Women's choices were analysed using a generalized multinomial logit model (G-MNL) and model estimates were used to predict product shares for each information condition. The predictions indicated that adverse information did not affect women's preferences for products relative to only receiving basic information. The promotional material increased women's preferences for the transdermal patch. Women in all groups had a low preference for the vaginal ring which was not improved by promotion. The findings highlight the need for researchers to pay attention to setting the context when conducting DCEs as this can significantly affect results.
Demirkale, F., Donovan, D.M. & Street, D. 2013, 'Constructing D-optimal Symmetric Stated Preference Discrete Choice Experiments', Journal of Statistical Planning and Inference, vol. 143, no. 8, pp. 1380-1391.
View/Download from: OPUS | Publisher's site
We give new constructions for DCEs in which all attributes have the same number of levels. These constructions use several combinatorial structures, such as orthogonal arrays, balanced incomplete block designs and Hadamard matrices. If we assume that only the main effects of the attributes are to be used to explain the results and that all attribute level combinations are equally attractive, we show that the constructed DCEs are D-optimal.
Knox, S.A., Viney, R.C., Street, D., Haas, M.R., Fiebig, D.G., Weisberg, E. & Bateson, D. 2012, 'What's good and bad about contraceptive products? A best-worst attribute experiment comparing the values of women consumers and GPs', Pharmacoeconomics, vol. 30, no. 12, pp. 1187-1202.
View/Download from: OPUS |
Our objectives were to compare women's (consumers') preferences and GPs' (providers') views in relation to existing and new contraceptive methods, and particularly to examine what factors increase the acceptability of different contraceptive products. Women and GPs agree that longer-acting methods with less bleeding are important features in preferred methods of contraception; however, women are also attracted to products involving less invasive modes of administration. While the vaginal ring may fill the niche in Australia for a relatively non-invasive, moderately long-acting and effective contraceptive, the results of this study indicate that GPs will need to promote the benefits of the vaginal ring to overcome negative perceptions about this method among women who may benefit from using it.
Street, D. & Knox, S.A. 2012, 'Designing for Attribute-Level Best-Worst Choice Experiments', Journal of Statistical Theory and Practice, vol. 6, pp. 363-375.
View/Download from: OPUS | Publisher's site
Using the D-optimality criterion, we show that resolution 3 fractional factorial designs perform as well as the complete factorial design in attribute-level best-worst choice experiments, assuming both that all attribute levels are equally attractive and that only main effects of attribute levels are to be used to explain the results.
King, M., Viney, R.C., Smith, D.P., Hossain, I., Street, D., Savage, E.J., Fowler, S., Berry, M.P., Stockler, M.R., Cozzi, P., Stricker, P.D., Ward, J. & Armstrong, B. 2012, 'Survival gains needed to offset persistent adverse treatment effects in localised prostate cancer', British Journal Of Cancer, vol. 106, no. 4, pp. 638-645.
View/Download from: OPUS | Publisher's site
Men diagnosed with localised prostate cancer (LPC) face difficult choices between treatment options that can cause persistent problems with sexual, urinary and bowel function. Controlled trial evidence about the survival benefits of the full range of treatment alternatives is limited, and patients' views on the survival gains that might justify these problems have not been quantified. A discrete choice experiment (DCE) was administered in a random subsample (n = 357, stratified by treatment) of a population-based sample (n = 1381) of men, recurrence-free 3 years after diagnosis of LPC, and 65 age-matched controls (without prostate cancer). Survival gains needed to justify persistent problems were estimated by substituting side effect and survival parameters from the DCE into an equation for compensating variation (adapted from welfare economics). Median (2.5, 97.5 centiles) survival benefits needed to justify severe erectile dysfunction and severe loss of libido were 4.0 (3.4, 4.6) and 5.0 (4.9, 5.2) months. These problems were common, particularly after androgen deprivation therapy (ADT): 40 and 41% overall (n = 1381) and 88 and 78% in the ADT group (n = 33). Urinary leakage (most prevalent after radical prostatectomy (n - 839, mild 41%, severe 18%)) needed 4.2 (4.1, 4.3) and 27.7 (26.9, 28.5) months survival benefit, respectively. Mild bowel problems (most prevalent (30%) after external beam radiotherapy (n = 106)) needed 6.2 (6.1, 6.4) months survival benefit.
Bush, S.A., Street, D. & Burgess, L.B. 2012, 'Optimal designs for stated choice experiments that incorporate position effects', Communications in Statistics - Theory and Methods, vol. 41, no. 10, pp. 1771-1795.
View/Download from: OPUS | Publisher's site
Davidson and Beaver (1977) extended the Bradley-Terry model to incorporate the possible effect of position within a choice set on the choices made in paired comparisons experiments. We further extend the Davidson-Beaver result to choice sets of any size and show, under a mild restriction, that designs optimal for the multinomial logit model are still optimal. Designs balanced for carry-over effects of all orders can be used to construct designs with a diagonal information matrix for attribute effects. The theoretical results are derived assuming equal merits and we discuss the possible consequences of assuming unequal merits in an example.
Fiebig, D.G., Knox, S.A., Viney, R.C., Haas, M.R. & Street, D. 2011, 'Preferences for new and existing contraceptive products', Health Economics, vol. 20, no. S1, pp. 35-52.
View/Download from: OPUS |
New contraceptive methods provide greater choice in terms of effectiveness, management of side-effects, convenience and frequency of administration and flexibility, but make the decisions about contraception more complex. There are limited data on the factors that determine women++s choices among these alternatives, to inform providers about the factors which are most important to women, or to predict uptake of new products. This paper reports on a choice experiment designed to elicit women++s preferences in relation to prescribed contraception and to forecast the impact of the introduction of two new products into the Australian market. A generalised multinomial logit model is estimated and used in the simulation exercise. The model forecasts that the hormonal patch would be well received among women, achieving a greater market share than current non-pill products, but the vaginal ring would have limited appeal.
Viney, R.C., Norman, R., King, M.T., Cronin, P.A., Street, D., Knox, S.A. & Ratcliffe, J. 2011, 'Time trade-off derived EQ-5D weights for Australia', Value in Health, vol. 14, no. 6, pp. 928-936.
View/Download from: OPUS | Publisher's site
Background: Cost-utility analyses (CUAs) are increasingly common in Australia. The EuroQol five-dimensional (EQ-5D) questionnaire is one of the most widely used generic preference-based instruments for measuring health-related quality of life for the estimation of quality-adjusted life years within a CUA. There is evidence that valuations of health states vary across countries, but Australian weights have not previously been developed. Methods: Conventionally, weights are derived by applying the time trade-off elicitation method to a subset of the EQ-5D health states. Using a larger set of directly valued health states than in previous studies, time trade-off valuations were collected from a representative sample of the Australian general population (n = 417). A range of models were estimated and compared as a basis for generating an Australian algorithm. Results: The Australia-specific EQ-5D values generated were similar to those previously produced for a range of other countries, but the number of directly valued states allowed inclusion of more interaction effects, which increased the divergence between Australia's algorithm and other algorithms in the literature. Conclusion: This new algorithm will enable the Australian community values to be reflected in future economic evaluations.
Burgess, L.B., Street, D. & Wasi, N. 2011, 'Comparing designs for choice experiments: A case study', Journal of Statistical Theory and Practice, vol. 5, no. 1, pp. 25-46.
View/Download from: OPUS | Publisher's site
This paper describes an empirical comparison of the performance of four designs for a discrete choice experiment. These designs were chosen to represent the range of construction techniques that are currently popular for choice experiments when no prior knowledge of the parameters is available. Each design had 320 respondents who each completed 16 choice sets. The results suggest that for the multinomial logit model (MNL) the design that is used at this stage is fairly unimportant. As the sample size gets smaller, however, differences between the designs become apparent. We also analysed the results using four different models which accommodate preference heterogeneity. We find that any of these models are able to predict choices more accurately for born in-sample and out-of-sample than the MNL model for the designs used here, and that the differences across designs arc larger for models with more parameters, although preliminary results suggest the gain appears to depend on the underlying preference structure.
Norman, R., King, M.T., Clarke, D., Viney, R.C., Cronin, P.A. & Street, D. 2010, 'Does mode of administration matter? Comparison of online and face-to-face administration of a time trade-off task', Quality of Life Research, vol. 19, no. 4, pp. 499-508.
View/Download from: OPUS | Publisher's site
Purpose Developments in electronic data collection methods have allowed researchers to generate larger datasets at lower costs, but relatively little is known about the comparative performance of the new methods. This paper considers the comparability of two modes of administration (face-to-face and remote electronic) for the time trade-off. Method Data were collected from a convenience sample of adults (n = 135) randomised to either a face-to-face time trade-off or a remote electronic tool. Patterns of responses were considered. For each sample, standard regression analysis was undertaken to generate a valuation set, which were then contrasted
Bush, S.A., Burgess, L.B. & Street, D. 2010, 'Optimal designs for stated choice experiments that incorporate ties', Journal of Statistical Planning and Inference, vol. 140, no. 7, pp. 1712-1718.
View/Download from: OPUS | Publisher's site
In 1970 Davidson generalised the Bradley-Terry model to allow respondents to say that the two options presented in a choice task were equally attractive. In this paper we extend this idea to the MNL model with m options in each choice set and we show that the optimal designs for the MNL model are also optimal in this setting.
Fiebig, D.G., Haas, M.R., Hossain, I., Street, D. & Viney, R.C. 2009, 'Decisions about Pap tests: What influences women and providers?', Social Science & Medicine, vol. 68, no. 10, pp. 1766-1774.
View/Download from: OPUS | Publisher's site
Despite the success internationally of cervical screening programs debate continues about optimal program design. This includes increasing participation rates among under-screened women, reducing unnecessary early re-screening, improving accuracy of and confidence in screening tests, and determining the cost-effectiveness of program parameters, such as type of screening test, screening interval and target group. For all these issues, information about consumer and provider preferences and insight into the potential impact of any change to program design on consumer and provider behaviour are essential inputs into evidence-based health policy decision making. This paper reports the results of discrete choice experiments to investigate women's choices and providers' recommendations in relation to cervical screening in Australia. Separate experiments were conducted with women and general practitioners, with attributes selected to allow for investigation of how women and general practitioners differ in their preferences for attributes of screening programs. Our results indicate a considerable commonality in preferences but the alignment was not complete. Women put relatively more weight on cost, chance of a false positive and if the recommended screening interval were changed to one year.
Norman, R., Cronin, P.A., Viney, R.C., King, M.T., Street, D. & Ratcliffe, J. 2009, 'International comparisons in valuing EQ-5D health states: A review and analysis', Value in Health, vol. 12, no. 8, pp. 1194-1200.
View/Download from: OPUS | Publisher's site
Objective: To identify the key methodological issues in the construction of population-level EuroQol 5-dimensions (EQ-5D)/time trade-off (TTO) preference elicitation studies. Method: This study involved three components. The first was to identify existing population-level EQ-5D TTO studies. The second was to illustrate and discuss the key areas of divergence between studies, including the international comparison of tariffs. The third was to portray the relative merits of each of the approaches and to compare the results of studies across countries. Results: While most articles report use of the protocol developed in the original UK study, we identified three key areas of divergence in the construction and analysis of surveys. These are the number of health states valued to determine the algorithm for estimating all health states, the approach to valuing states worse than immediate death, and the choice of algorithm. The evidence on international comparisons suggests differences between countries although it is difficult to disentangle differences in cultural attitudes with random error and differences as a result of methodological divergence. Conclusions: Differences in methods may obscure true differences in values between countries. Nevertheless, population-specific valuation sets for countries engaging in economic evaluation would better reflect cultural differences and are therefore more likely to accurately represent societal attitudes.
Louviere, J.J., Islam, T., Wasi, N., Street, D. & Burgess, L.B. 2008, 'Designing Discrete Choice Experiments: Do Optimal Designs Come At A Price?', Journal of Consumer Research, vol. 35, no. 2, pp. 360-375.
View/Download from: OPUS | Publisher's site
In discrete choice experiments, design decisions are crucial for determining data quality and costs. While high statistical efficiency designs are desirable, they may come at a price if they increase the cognitive burden for respondents. We address this problem by designing 44 experiments that systematically vary numbers of attributes and attribute level differences. Our results for two product categories suggest that respondents systematically are less consistent in answering choice questions as statistical efficiency increases. This relationship holds regardless of the number of attributes and is statistically significant even if one accommodates preference heterogeneity. Implications for practice and future research are discussed.
Louviere, J.J., Street, D., Burgess, L.B., Wasi, N., Islam, T. & Marley, A.A. 2008, 'Modeling the choices of individual decision-makers by combining efficient choice experiment designs with extra preference information', Journal of Choice Modelling, vol. 1, no. 1, pp. 128-163.
View/Download from: OPUS | Publisher's site
Street, D. & Burgess, L.B. 2008, 'Some open combinatorial problems in the design of stated choice experiments', Discrete Mathematics, vol. 308, no. 13, pp. 2781-2788.
View/Download from: OPUS | Publisher's site
Stated choice experiments are widely used in many areas and the optimal allocation of options to choice sets can make a substantial difference to the cost of running such an experiment. In this paper we describe some open problems in the design of optima
Gerard, K., Salisbury, C., Street, D., Pope, C. & Baxter, H. 2008, 'Is fast access to general practice all that should matter? A discrete choice experiment of patients' preferences', Journal of Health Services Research and Policy, vol. 13, no. 2, pp. 3-10.
View/Download from: OPUS
Objectives: To determine the relative importance of factors that influence patient choice in the booking of general practice appointments for two health problems. Methods: Two discrete choice experiments incorporated into a survey of general practice patients and qualitative methods to support survey development. Results: An overall response of 94% (1052/1123) was achieved. Factors influencing the average respondent's choice of appointment, in order of importance, were: seeing a doctor of choice; booking at a convenient time of day; seeing any available doctor; and having an appointment sooner rather than later (acute, low worry condition). This finding was the same for an ongoing, high worry condition but in addition the duration of the appointment was also of (small) value. Patients traded off speed of access for more convenient appointment times (a willingness to wait an extra 2.5-3 days longer to get a convenient time slot for an acute low worry/ongoing, high worry condition, respectively). However, contrary to expectation, patients were willing to trade off speed of access for continuity of care (e.g. willingness to wait five days longer to see the doctor of their choice for an acute, low worry condition). Preferences varied by a person's gender, work and carer status. Conclusions: Patients hold strong preferences for the way general practice appointment systems are managed. Contrary to current policy on improving access to primary care patients value a more complex mix of factors than fast access at all costs. It is important that policy-makers and practices take note of these preferences.
Burgess, L.B. & Street, D. 2006, 'The optimal size of choice sets in choice experiments', Statistics, vol. 40, no. 6, pp. 507-515.
View/Download from: OPUS | Publisher's site
Inthis paper, we establish the optimal size of the choice sets in generic choice experiments for asymmetric attributes when estimating main effects only. We give an upper bound for the determinant of the information matrix when estimating main effects and all two-factor interactions for binary attributes. We also derive the information matrix for a choice experiment in which the choice sets are of different sizes and use this to determine the optimal sizes for the choice sets.
Burgess, L.B. & Street, D. 2005, 'Optimal designs for choice experiments with asymmetric attributes', Journal Of Statistical Planning And Inference, vol. 134, no. 1, pp. 288-301.
View/Download from: OPUS | Publisher's site
In this paper we establish the form of the optimal design for choice experiments in which attributes need not have the same number of levels for testing main effects only, when there are k attributes, and all choice sets are of size m. We give a construc
Street, D., Burgess, L.B. & Louviere, J.J. 2005, 'Quick and easy choice sets: Constructing optimal and nearly optimal stated choice experiments', International Journal of Research in Marketing, vol. 22, no. 4, pp. 459-470.
View/Download from: OPUS | Publisher's site
In this paper we compare a number of common strategies for constructing discrete choice experiments. Two of the strategies, including one based on theoretical constructions for optimal discrete choice experiments, produce designs that are better than those that come about from random grouping and from using the LMA construction. A simple account of this theoretical construction is given
Street, D. & Burgess, L.B. 2004, 'Optimal and near-optimal pairs for the estimation of effects in 2-levels choice experiments', Journal of Statistical Planning and Inference, vol. 118, pp. 185-199.
View/Download from: OPUS | Publisher's site
This paper gives constructions for optimal and near-optimal sets of pairs for the estimation of main effects, and for the estimation of main effects and two factor interactions, in forced choice experiments in which all attributes have two levels. The number of pairs inthe sets is much smaller than that in previulsy constructed optimal 2-level choice experiments
Street, D. & Burgess, L.B. 2004, 'Optimal stated preference choice experiments when all choice sets contain a specific option', Statistical Methodology, vol. 1, pp. 37-45.
View/Download from: OPUS | Publisher's site
Stated preference choice experiments are routinely used in many areas from marketing to medicine. While results on the optimal choice sets to present for the forsed choice setting have been determined in a variety of situations, no results have appeared to date on the optimal choice sets to use when either all choice sets are to contain a common base alternative or when all choice sets contain a "none of these" option. These problems are considered in this paper.
Burgess, L.B. & Street, D. 2003, 'Optimal designs for 2(k) choice experiments', Communications In Statistics-theory And Methods, vol. 32, no. 11, pp. 2185-2206.
View/Download from: OPUS | Publisher's site
Smith, N.F. & Street, D. 2003, 'The use of balanced incomplete block designs in designing randomized response surveys', Australian & New Zealand Journal Of Statistics, vol. 45, no. 2, pp. 181-194.
View/Download from: OPUS | Publisher's site
Louviere, J.J., Street, D., Carson, R., Ainslie, A., Deshazo, J.R., Cameron, T.A., Hensher, D., Kohn, R. & Marley, A.A. 2002, 'Dissecting the random component of utility', Marketing Letters, vol. 13, no. 3, pp. 177-193.
View/Download from: OPUS | Publisher's site
Prvan, T. & Street, D. 2002, 'An annotated bibliography of application papers using certain classes of fractional factorial and related designs', Journal of Statistical Planning and Inference, vol. 106, no. N/A, pp. 245-269.
View/Download from: OPUS
In this paper we provide an annotated bibliography of about 140 papers which have appeared in journals in a variety of areas in the last 5 years and in which a fractional factorial design has been used. For each reference, we have indicated the design or designs used and whether or not the responses are given in the paper. The level of detail given in these papers makes them suitable sources for examples in a course on the design and analysis of experiments.
Street, D., Bunch, D.S. & Moore, B.J. 2001, 'Optimal Designs for 2k Paired Comparison Experiments', Communications in Statistics - Theory and Methods, vol. 30, no. 10, pp. 2149-2171.
View/Download from: OPUS | Publisher's site
In this paper we establish the form of the optimal paired comparison design when there are k attributes, each with two levels, for testing for main effects, for main effects and two factor interactions and for main effects and two and three factor interactions. In all cases we assume that all pairs with the same number of attributes different appear equally often. In this setting the D and A optimal designs for main effects are the foldover pairs and those for main effects and two factor interactions have pairs in which about half the attributes are different.
Street, D. & Wilson, W.H. 2001, 'Resolvable Designs for Resolving Disputes', Journal of Combinatorial Mathematics and Combinatorial Computing, vol. 38, pp. 139-148.
View/Download from: OPUS
Wang, I.Y., Fraser, I.S., Bursamian, S.P., Manconi, F., Street, D., Cornillie, F.J. & Russell, P. 2000, 'Endometrial Iysosomal enzyme activity in ovulatory dysfunctional uterine bleeding, IUCD users and post-partum women', Molecular Human Reproduction, vol. 6, no. 3, pp. 258-263.
View/Download from: OPUS | Publisher's site
The aim of this study was to evaluate the role of lysosomal enzymes in excessively heavy menstruation by comparing women with menorrhagia due to dysfunctional bleeding or intrauterine contraceptive device (IUCD) use with those with normal menstrual periods or with amenorrhoea associated with breastfeeding. This was a prospective cohort investigation of the activity of four endometrial lysosomal enzymes in three contrasting groups: (i) women with ovulatory dysfunctional uterine bleeding and users of intrauterine contraceptive devices; (ii) breastfeeding post-partum women in whom there are long periods of amenorrhoea, particularly in the early months post-partum; and (iii) normal cycling women. It was found that the total activity of lysosomal enzymes, particularly acid phosphatase and N-acetyl--D-glucosaminidase, was markedly elevated (P < 0.001) in IUCD-exposed endometrium, and endometrium from women with dysfunctional uterine bleeding when compared with endometrium from women with a history of entirely normal menstrual periods or that in post-partum breastfeeding women.

Other research activity

Louviere, J.J. & Street, D. 2003, 'Modelling the choices of individuals', ARC Discovery Grant.
Louviere, J.J. & Street, D. 2002, 'Modelling the choices of individuals', ARC.
Louviere, J.J. & Street, D. 2002, 'A rigorous theoretical and empirical examination of the properties of certain classes of experimental designs commonly used in choice modelling', ARC.