Descriptive analysis of responses to items in questionnaires. Why not using a fuzzy rating scale?

The Dirichlet Dual Response Model: An Item Response Model for Continuous Bounded Interval Responses

Standard response formats such as rating or visual analogue scales require respondents to condense distributions of latent states or behaviors into a single value. Whereas this is suitable to measure central tendency, it neglects the variance of distributions. As a remedy, variability may be measured using interval-response formats, more specifically the dual-range slider (RS2). Given the lack of an appropriate item response model for the RS2, we develop the Dirichlet dual response model (DDRM), an extension of the beta response model (BRM; Noel & Dauvier in Appl Psychol Meas, 31:47-73, 2007). We evaluate the DDRM's performance by assessing parameter recovery in a simulation study. Results indicate overall good parameter recovery, although parameters concerning interval width (which reflect variability in behavior or states) perform worse than parameters concerning central tendency. We also test the model empirically by jointly fitting the BRM and the DDRM to single-range slider (RS1) and RS2 responses for two Extraversion scales. While the DDRM has an acceptable fit, it shows some misfit regarding the RS2 interval widths. Nonetheless, the model indicates substantial differences between respondents concerning variability in behavior. High correlations between person parameters of the BRM and DDRM suggest convergent validity between the RS1 and the RS2 interval location. Both the simulation and the empirical study demonstrate that the latent parameter space of the DDRM addresses an important issue of the RS2 response format, namely, the scale-inherent interdependence of interval location and interval width (i.e., intervals at the boundaries are necessarily smaller).

Meat Analogues: Relating Structure to Texture and Sensory Perception

The transition from animal to plant proteins is booming, and the development of meat analogues or alternatives quickly progressing. However, the acceptance of meat analogues by consumers is still limited, mainly due to disappointing organoleptic properties of these foods. The objective of this study was to investigate possible relationships among structure, textural characteristics, consumer acceptance, and sensory evaluation of commercially available meat analogues. The microstructure and texture of 13 chicken analogue pieces and 14 analogue burgers were evaluated with confocal laser scanning microscopy (CLSM) and texture profile analysis (TPA). The moisture of the samples was related to cooking losses and release of liquid upon compression after cooking. Meat products were included as references. A sensory panel (n = 71) evaluated both flavour and texture characteristics. For the chicken analogue pieces, samples with more added fibres had a harder and chewier texture but were less cohesive. No other relations between composition and structure/texture could be found. In the sensory evaluation, lower hardness and chewiness were only seen in products with more fat. A lower sensory hardness was found to be related to the presence of small air pockets. For analogue burgers, there was no clear relation between composition and structure/texture. However, instrumentally measured hardness, chewiness, and cohesiveness correlated well with the corresponding sensory attributes, even though they could not be clearly linked to a structural feature. Next to this, fat content showed a clear correlation to perceived fattiness. CLSM images of burgers with high perceived fattiness showed large areas of fat. Therefore, the release of large fat pools from the meat was most likely responsible for the perception of this attribute. However, perceived fattiness was not related to liking, which was the case also for chicken analogue pieces. For both pieces and burgers, even if some of the measured textural attributes could be linked to the sensory profile, the textural attributes in question could not explain the liking scores. Liking was related to other aspects, such as meaty flavour and juiciness, which were not directly linked to compositional or textural features. Juiciness was not directly related to the moisture loss of the products, indicating that this attribute is rather complex and probably involves a combination of characteristics. These results show that to increase the appreciation of meat analogues by consumers, improving simple texture attributes is not sufficient. Controlling sensory attributes with complex cross-modal perception is probably more important.

Conditions Influencing Salary of the Automotive Industry in Mexico City—A Linguistic Fuzzy-Set Approach

Decision making in wages is generally a hard task. The aim of this work is to identify government conditions, personal conditions of the businessperson, and organizational circumstances that affect wage levels in the automotive industry in Mexico City using a linguistic fuzzy-set approach. We conducted a questionnaire, consisting of 23 observation variables with a five-point Likert scale. Independent variables were measured from 1 (“not important”) to 5 (“very important”). Based on the literature review and results of interviews, a total of 169 questionnaires were sent to participants using Google Forms. The results of the linguistic fuzzy-set approach identify three main conditions influencing the salary levels in the automotive industry in Mexico City, including unskilled manpower, the neoliberal economic model, and political and trade reforms. On the other hand, organizational conditions are not considered relevant in determining wage levels. Based on the findings, some recommendations have been proposed to help government, firm leaders, and businesspeople design appropriate personnel policies to achieve better salary satisfaction for employees in the future. This work shows a model based on the fuzzy-set approach that is a potential tool to overcome the difficulties posed by a complex environment.

Modeling random and non-random decision uncertainty in ratings data: a fuzzy beta model

Modeling human ratings data subject to raters’ decision uncertainty is an attractive problem in applied statistics. In view of the complex interplay between emotion and decision making in rating processes, final raters’ choices seldom reflect the true underlying raters’ responses. Rather, they are imprecisely observed in the sense that they are subject to a non-random component of uncertainty, namely the decision uncertainty. The purpose of this article is to illustrate a statistical approach to analyse ratings data which integrates both random and non-random components of the rating process. In particular, beta fuzzy numbers are used to model raters’ non-random decision uncertainty and a variable dispersion beta linear model is instead adopted to model the random counterpart of rating responses. The main idea is to quantify characteristics of latent and non-fuzzy rating responses by means of random observations subject to fuzziness. To do so, a fuzzy version of the Expectation–Maximization algorithm is adopted to both estimate model’s parameters and compute their standard errors. Finally, the characteristics of the proposed fuzzy beta model are investigated by means of a simulation study as well as two case studies from behavioral and social contexts.

Capturing richer information: On establishing the validity of an interval-valued survey response mode

Obtaining quantitative survey responses that are both accurate and informative is crucial to a wide range of fields. Traditional and ubiquitous response formats such as Likert and visual analogue scales require condensation of responses into discrete or point values—but sometimes a range of options may better represent the correct answer. In this paper, we propose an efficient interval-valued response mode, whereby responses are made by marking an ellipse along a continuous scale. We discuss its potential to capture and quantify valuable information that would be lost using conventional approaches, while preserving a high degree of response efficiency. The information captured by the response interval may represent a possible response range—i.e., a conjunctive set, such as the real numbers between 3 and 6. Alternatively, it may reflect uncertainty in respect to a distinct response—i.e., a disjunctive set, such as a confidence interval. We then report a validation study, utilizing our recently introduced open-source software (DECSYS), to explore how interval-valued survey responses reflect experimental manipulations of several factors hypothesised to influence interval width, across multiple contexts. Results consistently indicate that respondents used interval widths effectively, and subjective participant feedback was also positive. We present this as initial empirical evidence for the efficacy and value of interval-valued response capture. Interestingly, our results also provide insight into respondents’ reasoning about the different aforementioned types of intervals—we replicate a tendency towards overconfidence for those representing epistemic uncertainty (i.e., disjunctive sets), but find intervals representing inherent range (i.e., conjunctive sets) to be well-calibrated.

Gendered Beliefs in STEM Undergraduates: A Comparative Analysis of Fuzzy Rating versus Likert Scales

Women are underrepresented in growing positions such as those related to STEM field careers (i.e., science, technology, engineering, and mathematics). One of the causes for remaining out of that field could lie on gender stereotypes. Undergraduate stereotypes and beliefs are important as could easily uphold future gender segregation at the workplace. In the research arena the measurement of those biased beliefs is important as most commonly used Likert scales (LS) could raise problems in terms of accuracy. As fuzzy rating scales (FRS) are a promising measurement alternative, the aim of this study is to compare the properties of FRS against LS. We conducted a cross-sectional study with 262 STEM and non-STEM participants who answered to a questionnaire that, besides gendered beliefs and injustice perception towards the situation of women at the workplace, included personal characteristics as coursed degree and working experience. Results pointed out, on one hand, that FRS allowed for a better capture of the variability of individual responses, but on the other hand, that LS were better valued than FRS in what is concerned with satisfaction and ease of response. Advantages of FRS for psychosocial measurement are discussed to facilitate the study around causes of segregation that excludes women from the STEM labour market.

Are school psychologists' special education eligibility decisions reliable and unbiased?: A multi-study experimental investigation

Nearly 50 years of research show persistent racial disproportionality in the identification of special education disabilities, but the underlying mechanisms for these disparities remain largely unexplored. Because ambiguous regulations defining disabilities may allow subjectivity and unlawful differential treatment (i.e., racial bias or discrimination) in the special education eligibility process, an important target of study is disparate treatment of students by race in evaluations required to determine eligibility. School psychologists have long been recognized as highly influential in this process and in schools' resultant decisions. We used a 3 × 2 mixed factorial experimental design in three studies with simulated case report data to measure the influence of race and assessment data on school psychologists' perceptions of students' eligibility for special education in cases centering on emotional disturbance, intellectual disability, or autism, respectively. Participants included 302 practicing school psychologists in three states across the three experiments. There was little evidence of racial disparity, but participants tended to render decisions unsupported by, and even contrary to, evaluation data. Implications for research, practice, and professional development are discussed.