Two-stage Experimental Design Research Articles

Intra-household bargaining for a joint vacation

Taking a holiday trip is a common couple-based leisure activity in which both partners tend to be actively involved. This paper studies the intra-household bargaining for the choice of a vacation destination within couples. We conduct a discrete choice experiment in which we elicit both individual and couple preferences for different hypothetical travel portfolios in a two-stage experimental design. The couple choices are modelled as a function of males' and females' individual preferences, allowing for different bargaining weights for each characteristic of the holiday trip. Therefore, we assess partners’ bargaining power (influence) in the couple choices conditional on individual preferences. We find that although males have a more influential role overall, there seems to be a gender specialization in that females decide on the type of accommodation and males focus on the trip cost.

Valorization of agrowaste digestate via addition of wood ash, acidification, and nitrification

The valorization of low pollutant bioenergy waste streams as fertilizers is key for a more circular economy. In addition to supplementing nutrients, the wood fly ash could be used as stabilizing agent of an agrowaste digestate, to minimize the pollution of the environment. A two-stage experimental design with series of mild and severe acidifications was used to assess the impact on the availability of nitrogen, carbon, and phosphorus in the 25.26 ± 2.78 g of treated digestate. The 144-hour incubations at 20 °C and 100 rpm were done in closed chambers (i.e. 250-mL Schott Duran® bottles) with a 0.11 M sulfuric acid trap of 4.43 ± 0.12 mL. The proposed model for the interpretation of the results of the closed chamber indicated that the greatest rate of formation of ammoniacal nitrogen in the sulfuric fraction occurred during the first 48 h of incubation of the untreated PVWD (zero order kinetics: 2.31 ⋅ 10 −11 mol [H 2 SO 4 ] NH 4 + N/s). During this period, the mass transfer coefficient was 1.81 ⋅ 10 − 7 m/s , after which the equilibrium was established at 58.47 ± 20.50 mg [H 2 SO 4 ] NH 4 + N/kg digestate. Up to 14.08 ± 3.52 % of the water-soluble nitrogen was converted to nitric nitrogen in the remaining 96 h of incubation of the control. The acidification kept all the nitrogen in the form of ammonium but increased 5.97 ± 0.19 times the leachability of phosphorus. The combined treatment with wood ash and hydrochloric acid prevented both the nitrification and the increase of the share of water-soluble orthophosphate in the digestate. The high share of inert carbon in the wood fly ash was found promising to restore the soil as a natural carbon sink, upon application of the treated anaerobic digestate. Routes for the activation of ash as sorbent that would require lower doses of acid need to be investigated. • The greatest volatilization of ammonia was found in the untreated control digestate. • Closed chamber model was based on kinetic constant and mass transfer coefficient. • Regarding C and P, wood ash offered the greatest impact on the anaerobic digestate. • Wood ash acidification maintained ammoniacal nitrogen and low nutrient availability. • The addition of wood ash increased the N, C and P contents of the anaerobic digestate.

Synthesis and development of novel magnetic polymeric sorbent to simultaneous extraction of three anti-cancers using dispersive micro solid phase extraction procedure in biological and water samples

A suitable sample preparation method based on dispersive micro solid-phase extraction was developed to extract and per-concentrate three anti-cancers, including Imatinib (IMT), methotrexate (MTX), and irinotecan (IRT), before their determination using HPLC-UV. A novel magnetic sorbent containing an aluminum-metal organic framework (CAU-1), magnetic Fe3O4 nanoparticles, and a polymeric component (Polyaniline) was prepared to extract anti-cancers. The sorbent composite was optimized using an optimal mixture design. The optimum value of CAU-1, magnetic Fe3O4 nanoparticles and polyaniline in the sorbent composite was 0.528, 0.338, and 1.13 g, respectively. Factors influencing the IMT, MTX and IRT determination were optimized using a two-stage experimental design strategy. The method has limit of detections in the concentration range of 0.06–0.33 ng mL−1 for the IMT, MTX and IRT determination in water and urine samples. The linear ranges for the IMT, MTX and IRT determination were 0.3–256, 0.2–297, and 0.4–289 ng mL−1 with R2 of 0.9923, 0.9931, and 0.9918 in water samples, respectively. The intra and inter-day RSDs for the IMT, MTX and IRT determination were between 3.9 and 5.2 and 4.6–6.2 % for analyzing water and urine samples, respectively. The sorbent provided good reusability of six cycles of extraction/desorption procedure without a significant decrease in the extraction recovery toward the anti-cancers. The method’s ability was investigated by analyzing urine and real water samples, indicating that the method exhibited a proper recovery and RSD % in the range of 88.0–95.8 % and 5.8–7.3 %, respectively.

Efficient Sensor Calibration Via Machine Learning-Based Resampling Methods

It remains a challenge to accurately and precisely calibrate a sensor in an experimentally-efficient manner. The calibration task at hand is to quantify the functional relationship between the sensor’s response (denoted as y) and the target analyte concentration (denoted as x) based on laboratory experiments. In practical use, the sensor’s mathematical calibration model needs to be coupled with the device to make inference regarding the concentration of the analyte in an environment based on the observed sensor response. It is critical to provide a single-value estimate for the analyte concentration, as well as to quantify the uncertainty of the target estimate (i.e., to provide the possible range of the underlying concentration level).However, classic statistical methods are not able to provide valid inference on the target quantity of interest due to the following difficulties in sensor calibration. First, not only the typical forward modeling but also the inverse estimation is involved. The former refers to the generation of a calibration model functionally relating the analyte concentration to the sensor response (x→y) based on experimental data; The latter concerns the use of the calibration model in practical setting to obtain from the observed response the estimation of the unknown concentration (y→x). Second, the variability of the sensor responses tends to be heterogeneous as opposed to homogeneous, which is an assumption that classic statistical methods rely on.In light of the above, a computational resampling method was adapted in this work to address the machine learning (ML)-based inference issues (i.e., the uncertainty quantification of the estimated analyte concentrations). Built on the inference capability, a two-stage experimental design procedure was developed to guide the efficient sampling of calibration data via laboratory experiments.The ML-based inference and experimental design methods are integrated in the chemometric calibration procedure, which was applied on a simulated biosensor to demonstrate its efficiency over traditional methods. The sensor simulator is developed from the real experimental data for a paper-based lateral flow strip (PLFS) designed to measure in blood the diagnosis biomarkers for traumatic brain injury. PLFS is finding increasing applications in detection of various diseases in point-of-care testing settings. Figure 1

Two-stage Bayesian experimental design optimization for measuring soil–water characteristic curve

Two-stage Bayesian experimental design optimization for measuring soil–water characteristic curve

The effects of adaptation on maximum likelihood inference for nonlinear models with normal errors

This work studies the properties of the maximum likelihood estimator (MLE) of a multidimensional parameter in a nonlinear model with additive Gaussian errors. The observations are collected in a two-stage experimental design and are dependent because the second stage design is determined by the observations at the first stage. The MLE maximizes the total likelihood.Unlike most theory in the literature, the approximation made to the distribution of the MLE only involves taking the second stage sample size to infinity, as the resulting approximate model retains the dependency between stages, and therefore, more closely reflects the actual two-stage experiment.It is proved that the MLE is consistent and that its asymptotic distribution is a specific Gaussian mixture, via stable convergence. Finally, the efficiency of the adaptive procedure relative to the fixed procedure is illustrated by a simulation study under three parameter dose–response Emax and Exponential models.

The Effect of Democratically Determined (In)equality on Cooperative Behavior

We investigated the relationship between democratically determined economic inequality and cooperation in a two-stage experimental design. Although the relationship between inequality and cooperation has been studied extensively, experimental results in this area are contradictory and find inequality to have either a positive, negative, or no effect on cooperation. Our participants were randomly assigned to one of three societal classes (upper class, middle class, lower class) in a simulated small-scale society, and they subsequently voted to implement a societal system in which wealth was distributed either relatively equally or relatively unequally. We found lower levels of cooperation (measured as the invested amount in a public-good game) among societies that previously opted for the unequal distribution, but did not observe a general effect of a participant’s personal vote on cooperative behavior. Instead, middle-class and lower-class participants in unequal societies cooperated less than their counterparts in the equal societies, causing the observed differences on the societal level. These findings suggest that democratically induced policies that ultimately lead to greater equality of wealth are potentially able to have positive consequences on the readiness to cooperate on the production of public goods.

Assessing the effectiveness of gamification in reducing domestic energy consumption: Lessons learned from the EnerGAware project

The application of gamification to encourage energy conservation behaviour in house occupants is an emerging field of research. However, empirical evidence of its effectiveness is lacking. This paper presents lessons learnt from the EU-funded EnerGAware research project, in which an innovative serious game (a game designed for purposes other than purely entertainment) was developed to promote reduced energy consumption and carbon emissions by changing social housing tenants’ energy efficiency behaviour. The game was validated in a sample of European social housing using a longitudinal, two-stage experimental design, employing both pre-post and control group approaches. While some aspects of the game did not work as intended, there were nevertheless some positive impacts. The intervention increased social housing tenants’ awareness and engagement in certain energy saving behaviour and provided an average electricity saving of 3.46% and an average gas saving of 7.48%. Although savings were found not to be statistically significant, an effect size was detected (0.2). Therefore, future steps should exploit all available opportunities to replicate the pilot and increase the sample size so as to gain stronger evidence of the game's impact. Preliminary results support the utility of gaming investment in the household energy efficiency field, and provide useful insights and pathways that could be incorporated into the development of future serious game interventions to foster their effectiveness.

A two-stage experimental design for dilution assays.

Dilution assays to determine solute concentration have found wide use in biomedical research. Many dilution assays return imprecise concentration estimates because they are only done to orders of magnitude. Previous statistical work has focused on how to design efficient experiments that can return more precise estimates, however this work has not considered the practical difficulties of implementing these designs in the laboratory. We developed a two-stage experiment with a first stage that obtains an order of magnitude estimate and a second stage that concentrates effort on the most informative dilution to increase estimator precision. We show using simulations and an empirical example that the best two-stage experimental designs yield estimates that are remarkably more accurate than standard methods with equivalent effort. This work demonstrates how to utilize previous advances in experimental design in a manner consistent with current laboratory practice. We expect that multi-stage designs will prove to be useful for obtaining precise estimates with minimal experimental effort.

Are Individuals Luck Egalitarians? - An Experiment on the Influence of Brute and Option Luck on Social Preferences.

According to luck egalitarianism, inequalities should be deemed fair as long as they follow from individuals’ deliberate and fully informed choices (i.e., option luck) while inequalities should be deemed unfair if they follow from choices over which the individual has no control (i.e., brute luck). This study investigates if individuals’ fairness preferences correspond with the luck egalitarian fairness position. More specifically, in a laboratory experiment we test how individuals choose to redistribute gains and losses that stem from option luck compared to brute luck. A two-stage experimental design with real incentives was employed. We show that individuals (n = 226) change their action associated with re-allocation depending on the underlying conception of luck. Subjects in the brute luck treatment equalized outcomes to larger extent (p = 0.0069). Thus, subjects redistributed a larger amount to unlucky losers and a smaller amount to lucky winners compared to equivalent choices made in the option luck treatment. The effect is less pronounced when conducting the experiment with third-party dictators, indicating that there is some self-serving bias at play. We conclude that people have fairness preference not just for outcomes, but also for how those outcomes are reached. Our findings are potentially important for understanding the role citizens assign individual responsibility for life outcomes, i.e., health and wealth.

Development of a novel ultrasonic-assisted magnetic dispersive solid-phase microextraction method coupled with high performance liquid chromatography for determination of mirtazapine and its metabolites in human urine and water samples employing experimental design

Ultrasonic-assisted magnetic dispersive solid-phase microextraction coupled with high performance liquid chromatography has been developed for extraction and determination of mirtazapine, N-desmethyl mirtazapine, and 8-hydroxy mirtazapine in human urine and water samples. Magnetic graphene oxide-polyaniline nanocomposite (MGOPA) as a novel SPME sorbent was synthesized and used for the microextraction process. The analytical performance of MGOPA was compared with magnetic graphene oxide nanocomposite and indicated that the new sorbent was quite effective for extraction of mirtazapine, N-desmethyl mirtazapine, and 8-hydroxy mirtazapine. A two-stage experimental design approach, Plackett-Burman screening design and Box-Behnken optimization design, was used for screening and optimizing of significant variables in the microextraction process. The practical applicability of the proposed method was assessed by studying the linearity, intra-day and inter-day accuracy, enrichment factor, and precision. This method can be satisfactorily applied to the determination of mirtazapine and its metabolites in human urine and environmental water samples. Graphical Abstract Magnetic graphene oxide-polyaniline nanocomposite.

Search Strategies in Complex and Ambiguous Problem Spaces: Performance Comparisons Across Group Structures Through a Controlled Experiment

The effective use of group problem solving in the search for solutions over ambiguous problem spaces remains a challenge. Through a controlled two-stage experimental design we explore the impact of group structure, problem complexity, and search strategies on problem solving performance. We find support for prior theoretical predictions regarding the impact of problem complexity on the performance among different group structures. Our results also confirm that under certain problem complexity conditions collaborative groups may search more effectively. Through a granular examination, we demonstrate that the number of solutions generated during a search might be an imperfect proxy for performance: it is neither the only predictor of performance (solution space coverage and the distance of search increments matter), nor does it necessarily have a positive impact. Finally, the effects of diverse past solution attempts during search may shape collaborative search effectiveness in non-uniform ways.

A shrinking core model and empirical kinetic approaches in supercritical CO2 extraction of safflower seed oil

Utilization of supercritical CO2 in safflower seed extraction was performed using a semi-batch extractor. Different extraction parameters, such as 40–60MPa pressure, 323–347K temperature, 20–76min time, and 1–3mL/min CO2 flow rate were applied. A two-stage experimental design application was performed in order to maximize the oil yield. First of all, a 32 factorial design was applied to estimate the effect of the main factors and their interactions. The second part of the experimental design was improved and accelerated using the steepest ascent method. Optimum extraction parameters were determined to be 50MPa pressure, 347K temperature and 76min time at a constant CO2 flow rate (3mL/min) according to the 22 design. Under these conditions, the oil yield obtained was 39.42%, comparable with Soxhlet extraction (40%) for 8h. Shrinking core and empirical kinetic models were applied in order to generalize the extraction process. The predicted data was compatible with the experimental data.

An experimental investigation of reputation effects of disclosure in an investment/trust game

This paper examines experimentally the reputation building role of disclosure in an investment/trust game. It provides experimental evidence in support of sequential equilibrium behavior in a finitely repeated investment/trust game where information asymmetry raises the possibility of voluntary disclosure. I define two regimes, namely disclosure regime and no-disclosure regime and it is only in the disclosure regime that such disclosure of private information is a possibility. I compare investment levels across two regimes and find the startling result that investment is lower in disclosure regime. I find that this lower investment is attributable to the fact that the prior probability with which an investor in the disclosure regime believes that a manager is trustworthy is significantly lower than the prior probability with which an investor in the no-disclosure regime believes that a manager is trustworthy. I introduce a two-stage experimental design to homogenize prior beliefs about managers’ trustworthiness and find that after such homogenization, investment is higher in disclosure.

Two-Stage Experimental Design for Dose-Response Modeling in Toxicology Studies.

The efficient design of experiments (i.e., selection of experimental doses and allocation of animals) is important to establishing dose-response relationships in toxicology studies. The proposed procedure for design of experiments is distinct from those in the literature because it is able to adequately accommodate the special features of the dose-response data, which include non-normality, variance heterogeneity, possibly nonlinearity of the dose-response curve, and data scarcity. The design procedure is built in a sequential two-stage paradigm that allows for a learning process. In the first stage, preliminary experiments are performed to gain information regarding the underlying dose-response curve and variance structure. In the second stage, the prior information obtained from the previous stage is utilized to guide the second-stage experiments. An optimization algorithm is developed to search for the design of experiments that will lead to dose-response models of the highest quality. To evaluate model quality (or uncertainty), which is the basis of design optimization, a bootstrapping method is employed; unlike standard statistical methods, bootstrapping is not subject to restrictive assumptions such as normality or large sample sizes. The design procedure in this paper will help to reduce the experimental cost/time in toxicology studies and alleviate the sustainability concerns regarding the tremendous new materials and chemicals.

Reputation Effects of Disclosure: An Experimental Investigation

This paper examines experimentally the sequential equilibrium theory of reputation and disclosure developed in Lunawat, 2009. It provides experimental evidence in support of the sequential equilibrium play and of the reputation building role played by disclosure. The said paper predicts higher investment in economies with disclosure as compared to those without. In testing this prediction, this paper introduces a two-stage experimental design which turns out also to be a contribution to experimental methodology. The two-stage design allows for construction of subject samples with uniform prior beliefs about a manager’s trustworthiness but different disclosure institutions. Data collected using this design provides support for the said prediction. This paper also sheds light on a related question of the reputation building role played by incremental disclosure. It introduces a multi-period setting that allows reinvestment of capital and provides for definition of incremental disclosure in an economy. It is shown that incremental disclosure results in even higher investment.

Exploring anchoring effect and the moderating role of repeated anchor in electronic commerce

Most of the anchoring studies conduct the experiments following two-stage experimental design in which a comparative judgement is followed by an absolute estimate. However, this design is impractical in many situations, especially in online auction environments where consumers cannot be asked to make comparisons before they make estimates. This study suggests and validates an alternative experimental design where no comparison process is needed to induce anchoring bias. Moreover, we postulate that repeated occurrence of anchoring points affect the strength of the anchoring effect on participants' price estimates. Two experiments were conducted to investigate the anchoring effect in both two-stage and one-stage designs. The results showed that the anchoring effect was robust under two-stage design, and repeated occurrence of the anchor did not reinforce the anchoring effect. Moreover, people's price estimates are susceptible to the anchoring effect even when they were not asked to make a comparative judgement before they determine the final estimate. However, the anchoring effect can be further consolidated when the anchor appears three times rather than once. Experimental design introduced in study 2 provides a promising website design direction and we suggest that high anchors should be embedded in a web page repeatedly to influence the buyers' value estimates to produce a premium.

An Artificial Neural Network Based Metamodel for Analysing a Stochastic Combat Simulation

In this paper, we discuss the use of multi-layer perceptron (MLP) neural networks to predict the outputs from a stochastic combat simulation model. The use of a faster, auxiliary model to approximate the input-output relationships of a complex simulation model is known in the simulation community as metamodelling. Since many simulation models contain a large number of input parameters, however it is necessary to reduce the set considered and determine the most important ones to include in a metamodel given a particular modelling context. This paper employs a two-stage experimental design — first, Morris’ randomised one-at-a-time (OAT) design is used as a factor screening method, and then a Latin Hypercube design is employed to select the set of input configurations which defines the required runs of the simulation model. The set of input configurations together with the associated simulation outputs provide the training data for the development of the MLP networks. The approach is illustrated with reference to a stochastic combat simulation called SIMBAT. The paper then investigates a number of aspects relating to the development of neural network-based metamodels of stochastic combat simulations. It shows that using the outputs from each replication of a stochastic simulation is generally better than only using the mean output.

Bayesian application to the two-stage near-saturated experimental design method with dispersion effects

Experimental designs are critical in the quality improvement of manufacturing processes and to the development of new processes. Experimental methods are often used to identify the most important controllable variables to the product, and the relationship of the product response to the controllable variables can be determined by experimental data. Traditionally, regression analysis is used to establish an empirical mean model for the product response, and competing experimental designs are evaluated with regard to estimating coefficients in the regression model. However, most traditional procedures involve the assumption of homogeneous process variance, which in many processes is inappropriate. A two-stage experimental design procedure developed by Mays and Myers [Mays, D.P. and Myers, R.H., 1997, Design and analysis for a two-level factorial experiment in the presence of variance heterogeneity. Computational Statistics and Data Analysis, 26, 219–233.] has proven to be beneficial in many heterogeneous variance situations. In the procedure, the first stage estimates the heterogeneous variance structure, and the second stage augments the first stage design to create an efficient design for estimating the mean. However, the first stage variance estimation is not reliable for small first stage experiment sizes, and hence a modified Bayesian two-stage procedure has been proposed for such situations [Mays, D.P. and Myers, R.H., 1996, Bayesian approach for the design and analysis of a two level factorial experiment in the presence of dispersion effects. Communications in Statistics—Theory and Methods, 25, 1409–1428.]. This study applies the Bayesian two-stage procedure to Koshal designs, hybrid designs, and small composite designs and examines the variance estimation in the first stage in order to find the optimal number of replicates to make at each design location. The Bayesian procedure is compared with the non-Bayesian two-stage procedure and with two competing one-stage equal replicate procedures. The effects of the various variance information on the procedures are also investigated.

A typological model for estimation of drug and placebo effects in depression.

One reason for failure to find specific treatment effects in drug versus placebo trials with patients who have depression is an insufficient period of observation. Also, differentiating between early fleeting response and maintained response has been shown relevant to detecting specific drug action. A model in which various types of drug and placebo response are specifically stipulated and which takes advantage of a two-stage experimental design is proposed. Substantive findings when patients who have major depression with atypical features are studied are: 1) about 6% have only a fleeting response to placebo and no response to drug; 2) by week-10 about 28% of the population will respond even if no medication is given; and 3) specific response to imipramine (21%) can be determined by week-4 but specific response to fluoxetine (20%) cannot be determined until later (week-10 in this study).