Carla A. Vivacqua

Post-Fractionated Strip-Block Designs

Novel arrangements for strip-block designs that reduce the experimental effort are presented. Theoretical properties of strip-block designs using post-fractionation are provided, and appropriate data analysis is explained. An experiment on an industrial process is used as an illustration. As a tool to aid the selection of appropriate plans, catalogs of post-fractionated strip-block designs with 16 and 32 trials are provided.

Strip-Block Experiments for Process Improvement and Robustness

The execution of industrial experiments often involves two stages. For example, the experiment may involve two separate processing steps or, when conducting robustness experiments, it may be possible to separate the change of the design factors and the environmental factors into two separate steps. A third possibility occurs when some factors are hard to change and others relatively easy. In each of these cases, it is frequently possible to reduce the cost of experimentation substantially by using what in agriculture in the 1930s became known as stripblock experiments. In this column we explain what a strip-block experiment is and provide an illustrative industrial example.

Sexual dysfunction in obese women is more affected by psychological domains than that of non-obese

PURPOSE To compare differences in the occurrence and changed domains of sexual dysfunction in obese and non-obese Brazilian women. METHODS Female Sexual Function Index, based on six domains, to investigate 31 sexual dysfunction incidence for obese compared to 32 non-obese women, was used. Statistical analysis using ANOVA and MANOVA were performed to compare total scores of Female Sexual Function Index among groups and to identify the differences among domains, Student t -test was used. Statistical significant level was established for all tests for p<0.05. RESULTS No difference in female sexual dysfunction frequency between obese (25.8%) and non-obese women (22.5%) was found. However, an important distinction in which aspects of sexual life were affected was found. While the obese group was impaired in three domains of sexual life (desire, orgasm, and arousal), in the control group five aspects were dysfunctional (desire, orgasm, arousal, pain and lubrication). Future research exploring psychological outcomes in obese females, such as body image and measures of positive and negative effect, might better characterize the female sexual dysfunction in this group. CONCLUSIONS Obesity does not appear to be an independent factor for allow quality of female sexual life. However, disturbance associated to obesity indicates a low frequency of disorder in physical domains, suggesting that psychological factors seem to be mainly involved in the sexual dysfunction in obese women.

Annual variation in plasma cortisol levels in common marmosets, Callithrix jacchus

Abstract In contrast to numerous studies showing breeding seasonality for animals living in temperate zones, few have addressed the effect of annual environmental changes on hormones and behavior related to reproduction for animals living in tropical and equatorial regions. In order to expand information on other primates living near the equator, we used 13 adult common marmosets, eight males and five females living under natural climatic conditions in a colony located in North-Eastern Brazil. Blood samples were collected monthly from July 2002 to June 2004. Cortisol assays were performed by ELISA. Cortisol levels varied according to environmental temperature and rainfall but were not significantly affected by photoperiod. After separating out these environmental effects through an analysis of covariance for repeated measurements, a marginal significant difference in cortisol levels according to the seasons of the year was detected. A significant difference in cortisol levels between females and males was also observed. The results suggest that annual variation in cortisol in common marmosets is influenced by temperature and rainfall and might be associated with the mating season.

Minimum Setup Minimum Aberration Two‐level Split‐plot Type Designs for Physical Prototype Testing

Althoughnew technologies allow for less effortinprototyping, physical testing still remainsan importantstep in theproductdevelopment cycle. Well-planned experiments are useful to guide the decision-making process. During the design of anexperiment, one of the challenges is to balance limited resources and system constraints to obtain useful information. It iscommon that prototypes arecomposed of several parts,with some parts moredifficultto assemble thanothers. And, usually,there is only one piece available of each part type and a large number of different setups. Under these conditions, designswith randomization restrictions become attractive approaches. Considering this scenario, a new and additional criterion,minimum setup, to construct split-plot type designs is presented. Designs with the minimum number of setups of the moredifficult parts, which are especially useful for screening purposes in physical prototype testing, are discussed. The use of theproposed criterion combined with minimum aberration for selecting a regular design is shown through a real application intesting car prototypes. As a tool to practitioners, catalogs of selected 32-run minimum setup minimum aberration split-split-plot and split-split-split-plot designs are presented. More complete catalogs are available as Supporting information.Copyright

Not all Models are Polynomials

Data analysis is ultimately about making sense of a set of data—to discover systematic patterns that help predict, or nearly so, future outcomes. In the context of factorial experiments, most statistical textbooks give the impression that we must analyze the data with the aid of a factorial model. However, this is an unnecessarily dogmatic position. Mother nature is not obliged to behave factorially! A factorial model is not always the best way to summarize the result of an experiment. We demonstrate in this column, with an example from Brownlee (Brownlee, K. A. (1953). Industrial Experimentation. New York: Chemical Publishing Co. Inc.), that there are situations where a factorial polynomial model is not the best way to summarize the result of an experiment. This example also is an illustration of the power of data transformations.

Development and validation of an UHPLC method for the determination of betamethasone valerate in cream, gel, ointment and lotion

An ultra high performance liquid chromatographic method has been developed and validated for the determination of betamethasone valerate (BMV) in topical dermatologic formulations. For the development of the method, response surface methodology based on a three-level full factorial design was used. The eluent composition, the column dimension and the flow rate were chosen as relevant experimental parameters to investigate. The response surface plots revealed an optimum separation by using a RP column (30 mm × 2 mm i.d., 2.2 μm particle size), at 30 °C; isocratic mobile phase consisting of acetonitrile:water (60:40) at a flow rate of 0.2 mL min(-1) and a wavelength set at 254 nm. The proposed method was validated for four types of matrices according to ICH guidelines requirements. Dexamethasone acetate (DMA) was used as internal standard. Linearity was studied in the range of 5-200 μg mL(-1) for BMV in spiked matrix samples. Recoveries were in the range of 95-105% and precision was better than 5% for both analytes, either in cream, gel, ointment, or lotion formulations, when using simple sample preparation. Retention times were 0.95 min for DMA and 1.40 min for BMV, demonstrating a short method run time. The method was successfully applied for routine analysis of dermatological formulations containing betamethasone valerate.

Analysis of Augmented Unreplicated Factorial Designs Repeated in Time

Unreplicated designs are fairly common in industrial applications; however, there is resistance to their use in agricultural science. In the agriculture community, there is still a belief that lack of replication may prevent the experimenter from getting useful conclusions. Nevertheless, sound statistical methods that permit valid comparisons in unreplicated studies are available for many types of designs. The objective of this paper is to present an analysis procedure for unreplicated designs combining typical characteristics found in industrial experimentation (factorial designs augmented with center points) and in agricultural applications (inclusion of control treatments and repeated measurements). We illustrate the method through a real experiment to evaluate the use of sugarcane by-products in chicken diet. Specifically, it is an unreplicated two-level factorial design with two additional runs (a center point and a control treatment), with experimental units measured in two periods of time. Replication was initially planned in the case study, but the actual treatment application led to an unreplicated design. The application of the proposed method allows interpretation of the data collected. We conclude that the appropriate use of unreplicated designs in agricultural and biological research may reduce overall costs and lessen the use of in vivo testing. Copyright

Mixed two- and four-level experimental designs for interchangeable parts with different degrees of assembly difficulty

Purpose The purpose of this paper is to show how to properly use the method of replacement to construct mixed two- and four-level minimum setup split-plot type designs to accommodate the presence of hard-to-assemble parts. Design/methodology/approach Split-plot type designs are economical approaches in industrial experimentation. These types of designs are particularly useful for situations involving interchangeable parts with different degrees of assembly difficulties. Methodologies for designing and analyzing such experiments have advanced lately, especially for two-level designs. Practical needs may require the inclusion of factors with more than two levels. Here, the authors consider an experiment to improve the performance of a Baja car including two- and four-level factors. Findings The authors find that the direct use of the existing minimum setup maximum aberration (MSMA) catalogs for two-level split-plot type designs may lead to inappropriate designs (e.g. low resolution). The existing method of replacement for searching exclusive sets of the form (α, β, αβ) available in the literature is suitable for completely randomized designs, but it may not provide efficient plans for designs with restricted randomization. Originality/value The authors provide a general framework for the practitioners and have extended the algorithm to find out the number of generators and the number of base factor at each stratum, which guide the selection of mixed two-level and four-level MSMA split-plot type designs.