Comparison of statistical analysis methods for object case best–worst scaling

Abstract

Aims: Different methods have been used to analyze “object case” best–worst scaling (BWS). This study aims to compare the most common statistical analysis methods for object case BWS (i.e. the count analysis, multinomial logit, mixed logit, latent class analysis, and hierarchical Bayes estimation) and to analyze their potential advantages and limitations based on an applied example.

Methods: Data were analyzed using the five analysis methods. Ranking results were compared among the methods, and methods that take respondent heterogeneity into account were presented specifically. A BWS object case survey with 22 factors was used as a case study, tested among 136 policy-makers and HTA experts from the Netherlands, Germany, France, and the UK to assess the most important barriers to HTA usage.

Results: Overall, the five statistical methods yielded similar rankings, particularly in the extreme ends. Latent class analysis identified five clusters and the mixed logit model revealed significant preference heterogeneity for all, with the exception of three factors.

Limitations: The variety of software used to analyze BWS data may affect the results. Moreover, this study focuses solely on the comparison of different analysis methods for the BWS object case.

Conclusions: The most common statistical methods provide similar rankings of the factors. Therefore, for main preference elicitation, count analysis may be considered as a valid and simple first-choice approach. However, the latent class and mixed logit models reveal additional information: identifying latent segments and/or recognizing respondent heterogeneity.     

A best–worst scaling in Colombian patients to rank the characteristics of HIV/AIDS treatment

Aim: To elicit patients’ preferences for HIV/AIDS treatment characteristics in Colombia.

Materials and methods: A best–worst scaling case was used to provide a ranking of 26 HIV/AIDS treatment characteristics that were similar to a previous study conducted in Germany. In each choice task, participants were asked to choose the most important and the least important treatment characteristics from a set of five from the master list. Using the Hierarchical Bayes method, relative importance scores were calculated. Sub-group analyses were conducted according to sex, education, source of infection, symptoms, and age.

Results: A total of 195 patients fully completed the questionnaire. The three most important characteristics were “drug has very high efficacy” (relative importance score [RIS]?=?10.1), “maximum prolongation of life expectancy” (RIS?=?9.7), and “long duration of efficacy” (RIS?=?7.4). Sub-group analysis showed only three significant (but minor) differences between older and younger people.

Conclusion: This study suggests that treatment characteristics regarding efficacy and prolongation of life are particularly important for patients in Colombia. Further investigation on how patients make trade-offs between these important characteristics and incorporating this information in clinical and policy decision-making would be needed to improve adherence with HIV/AIDS medication.     

A new approach to determining sectoral priorities in an economy: input–Output elasticities

This paper uses input–Output elasticities to identify important economic sectors. Elasticities of output employment and income are used to identify key sectors of the Greek economy. A comparison of the rankings of economic sectors based on input–output elasticities with those based on net backward linkages indicates significant divergence in sectoral rankings obtained from the two approaches. The elasticity approach yields more consistent estimates of sectoral output employment and income potentials than the net backward linkage approach. Measured in terms of the potential to generate output employment and income agriculture services and textiles are found to be the key sectors for the Greek economy.     

Using the optimization algorithm to evaluate and predict the business performance of logistics companies–a case study in Vietnam

ABSTRACT

In the current market economy, it is important to evaluate and forecast the situation and business performance of enterprises to provide the necessary information for managers to plan for future use of resources. This study aims to evaluate and predict the business performance of logistics companies in Vietnam. The authors use the optimal algorithm in the data envelopment analysis (DEA) to evaluate the business efficiency of the companies in the years 2014–2017. In addition, the authors use Grey system theory to forecast business results and their future use during the period of 2018–2021. The research shows that Gemadept Corporation and Sea & Air Freight International use the their business resources effectively and as expected, and that these companies will continue to thrive in the future. This study provides a method to measure, evaluate, and forecast the business performance of the logistics companies. Managers and the government can rely on this approach for implementation and overall planning of logistics enterprises in the future.     

Ecosystem services: Foundations for a new rural–urban compact

The traditional rural–urban compact, now almost ten thousand years old, whereby the countryside sent products and people to the city in exchange for the city's products, services and governance is not working anymore. The rural population is increasingly marginalized and natural environments are increasingly destroyed. A new rural–urban compact needs to arise where cities acknowledge and pay for environmental sustainability. In this new rural–urban compact there would be more employment opportunities and more income coming to the rural areas, and the cities would benefit from a sustainable supply of rural products and ecosystem services provided by restored rural environments. This paper describes why this is important and provides several ideas on how to implement programs.     

Using a design structure matrix to support technology roadmapping for product–service systems

In light of the increasing trend of the use of the product–service system (PSS), tools that can help to analyse complex relationships among products, services and technologies and furthermore support decision-making during their planning are needed. To meet these needs, this study proposes a design structure matrix (DSM)-based technology roadmap (TRM) approach and a case study on mobile services is conducted. While PSS planning may involve a number of teams and organisations, the DSM-based TRM can help with such complex collaborative technology planning and hence lead to better performance. Methodologically, this is one of the first attempts to link DSM and TRM, both of which are widely used in planning stages, and thus is expected to contribute to the advances in TRM methodologies. In practice, this study describes an overall roadmapping process for the PSS, thus providing worthwhile guidelines available to organisations that are in charge of the TRM.     

Testing for unit roots: a simple alternative to Dickey—Fuller

This ariticle proposes a very simple test for a unit root in a time series which is based on the residual autocorrelation function from a regression of the observed series on a linear trend. The test statistic is proportional to the maximum lag length for which the residual autocorrelations at all lower lag lengths are stickly positive. Critical values for the test are calculated using Monte Carlo simultation assuming a random walk DGP. Its robustness properties to different I(1) DGPs are examined along with its power characteristics. These are found to compare very favourably to those of Dickey–Fuller-type tests.     

Private Remedies for Private Anti-Competitive Barriers to Trade: The Kodak–Fuji Example

The article describes a number of remedies available to a private US company for private anti-competitive activity in its export market, including action in US courts or the courts of the foreign market, and World Trade Organization (WTO) dispute resolution. The article describes the problems which a would-be plaintiff might perceive. One interesting result is that no matter how good the case on the merits, the plaintiff might prefer the most political avenue for relief.     

Immigrant earnings returns to post-migration education: Evidence for Canada, 1999–2013

Using the recently created Canadian Employer–Employee Dynamics Database (CEEDD), we analyze the earnings returns to investments in post secondary education by immigrants made after arrival in Canada. Fixed effects estimation results suggest that postsecondary education (PSE) in Canada increases annual earnings of men and women by 21.9% and 32.8%, respectively, for those enrolled full time and by 11.2% and 21.0% for those enrolled part time, respectively. Earnings effects of PSE are larger for immigrants admitted in the skilled worker category relative to family class immigrants and to refugees and are increasing in the level of education of the immigrants at landing. Immigrants from non-traditional source countries generally have larger returns to PSE than immigrants from English-language source countries, such as the US and the countries of Northwestern Europe.     

Doing a Monty: Who Opened the Door to This Game for Economists?

The Monty Hall three-door, "Let's Make a Deal" game, named after the 1970s television show, is used widely in economics, econometrics, statistics, and game-theory-based teaching, as well as in many other disciplines. Its solutions and underlying assumptions arouse great passion and argument, in both the academic and popular press. Most economists believe that the first popular elucidation of the game was presented by Nalebuff in the Journal of Economic Literature. However, the game and its controversial solution were spelled out in The American Statistician years earlier by Selvin, at the time a young biostatistician at the University of California at Berkeley. Mathematicians give him due credit. The author argues that economists should also recognise his contribution.     

Metaphors for manufacturing: What could it be like to be a manufacturing system?

In the world of modern manufacturing, considerable emphasis is placed upon properties of manufacturing systems described by the terms “flexibility,” “complexity,” “reliability,” “self-repairing,” and so forth. To understand and deal with such properties, one needs a theoretical framework allowing one to pose and analyze various questions surrounding the meaning and interrelations of these terms. This paper addresses the questions of what sort of system-theoretic frameworks would be likely candidates to form the basis for such a theory of manufacturing.Our approach to the modeling problem is to examine several different paradigms that have proven useful in other fields—engineering, biology, linguistics, computer science, chemistry—and to explore the degree to which these “metaphors” can be used to characterize manufacturing systems. The paper concludes with an assessment of the strengths and weaknesses of each metaphor and a suggestion for a program devoted to further development of the most promising approaches.     

Industrial invention: a supply and demand model for the UK, 1961–1989

A supply and demand model of industrial invention is presented paying particular attention to the impact of past advances on present advances in knowledge. The model is estimated using a recent span of UK data: a good fitting specification is found, which passes a battery of diagnostic tests, and possesses sensible long-run properties and an interpretable lag structure. The specific results found should be of potential interest to governments wishing to encourage the flow of inventions by fiscal means.     

Labor productivity forecasts based on a Beveridge–Nelson filter: Is there statistical evidence for a slowdown?

In this paper, I assess the evidence for a structural break in labor productivity growth in the years before the Great Recession with the use of out-of-sample forecasting exercises for the years 2010 to 2019 and the recently developed Beveridge–Nelson filter. Models based on a Beveridge–Nelson filter with no structural breaks outperform those allowing for a structural break, and there is statistically significant evidence that they outperform the random walk, though all models were too optimistic about labor productivity growth. Recently developed statistical tests do point to the presence of a structural break before the Great Recession, but uncertainty about the data-generating process for labor productivity growth or the timing and magnitude of the break may be too great to be helpful in forecast preparation.     

The need for a strategic approach on the railways to the management of the vehicle–track interface

Network Rail's decision in 2003 to bring track maintenance back in-house was long-overdue recognition that responsibility for safety on Britain's railways cannot be abdicated – but there was more to it than that. The accidents at Hatfield and Potter's Bar (both in Hertfordshire, UK), in particular, had exposed very vividly how privatisation had encouraged each now separated and competing part of the industry to operate in its own interests: records, data and expertise were fragmented, and track and train operators alike each sought to minimise their individual costs in isolation. This paper argues from a systems analysis perspective that, on vehicle and track maintenance at least, the overall costs to the industry are in such circumstances likely to be higher than if the various parties were to work together towards minimising total track and train maintenance costs between them. Far from increasing efficiency, privatisation had the effect of reducing it.     

Foresight in innovation policy: Shared visions for a science park and business–university links in a city region

Many car manufacturers recognize fuel cell vehicles as future substitutes for conventional cars with internal combustion engines. According to press releases and brochures, different strategic approaches of the automobile companies to fuel cell technology can be identified. These strategies match to a high degree the market entry strategies known from strategic marketing literature. A system dynamics model that reflects the beginning innovation process and the strategic approaches of a pioneer (first to market), an early follower (early to market) and a late follower (late to market) has been built. It examines the future prospects of the car manufacturers’ strategies in three different scenarios, which illuminate possible future developments of external influences like politics or fuel infrastructure.     

Internationalisation,growth and pay inequality: a cointegration analysis for Turkey, 1970–2007

The nature of the internationalisation-growth-inequality nexus is very complex; and therefore, not surprisingly, there is no consensus on whether increasing openness to trade leads to higher inequality or not – in fact, there is even no full understanding on the impact of the openness on the economic growth. In the case of Turkey it is observed that there is relatively little empirical work that addresses the issue of inequality. Considering this fact, the study aims to provide some more evidence on the complex relationship between trade openness, foreign direct investment, economic growth and pay inequality by utilising a combination of different Theil Indices of pay inequality. The paper utilises the Johansen Cointegration and Granger Causality methods. Our findings yield that higher economic growth resulting from trade openness comes with higher pay inequality.     

Public–private partnerships for the provision of public goods: Theory and an application to NGOs

When will a public good or service be provided by the government, when will it be provided by a NGO, and when will we see a private–public partnership? This paper provides a model where a typical public good requires different inputs which raises the possibility of partnerships to exploit comparative advantages of different parties. But hold-up problems due to contractual incompleteness in specifying tasks discourage separation of ownership and management. The fact that public goods have the property of non-rivalry and non-excludability and that NGOs tend to be non-profits drives our key results. We apply the framework to NGOs in developing countries which, in the last few decades, have been increasingly involved in various capacities in the provision of a wide range of public goods and services.     

Original Research A modified method for estimating volume–outcome relationships: application to percutaneous coronary intervention

Objective: The objective of the current study was to propose an alternative method for measuring individual operator and peer volumes to use as predictors for adverse outcomes.

Study design: A retrospective analysis was performed to assess the volume–outcome relationship for percutaneous coronary intervention (PCI) performed in New York State between 1996 and 1999. This relationship was calculated using a modified method whereby physician volume was calculated using the previous year's volume, and hospital volume was calculated after subtracting the operator of interest's annual volume from the total. The primary outcome of interest was in-hospital mortality.

Results: Using the modified method, the odds ratio (OR) of in-hospital mortality was 0.74 (95% confidence interval (CI) 0.55–0.99; p=0.04) for cardiologists who performed 75–174 procedures annually and 0.80 (95% CI 0.61–1.04; p=0.1) for cardiologists who performed ≥175 procedures annually compared with the lowest-volume operators. With the conventional approach to volume measurement, no relationship between cardiologist volume and mortality was found. Patients who underwent PCI in hospitals where their physician's peers had an annual volume of 600–999 or ≥1,000 cases had a significantly reduced odds of mortality (OR = 0.73; 95% CI 0.57–0.92; p=0.01; and OR = 0.77; 95% CI 0.62–0.95; p=0.01) compared with patients treated by physicians with an annual peer volume of <600 cases. The conventional method did not detect any significant correlation between hospital volume and in-hospital mortality.

Conclusion: The alternative approach to measuring cardiologist and peer volumes proposed in this study leads to more precise estimates of volume–outcome relationships than the conventional approach.     

The Higher Earning in America: Are 529 Plans a Good Way to Save for College?

This article studies whether 529 plans are an effective way for most people to save for college. The 529 plans were created in 1996 to help low- and middle-income American families save for college. Since this time they have adopted more tax advantages and grown substantially as a result. While total balances in 529 plans now exceed $250 billion, less than 3.0 percent of households have a 529 plan. And the majority of 529 plan accounts are held by the wealthiest households. Low- and middle-income households receive little tax savings from investing in 529 plans, and some states count these plans against financial aid. Also, 529 plans are complicated, rules vary by state, and fund management fees tend to be high. Thus, 529 plans are not the panacea to college affordability.