Human assessors are currently the best instruments for evaluating the sensory properties of both food and non-food consumer products. In the early stages of descriptive sensory panel training, a lexicon is still undergoing refinement and assessors are only beginning to understand the identities of sensory attributes and how to use the scale to communicate the intensity of sensations. To expedite training, the panel leader provides structure and feedback. Several statistical approaches have been proposed that indicate training status. In this thesis several approaches are reviewed, and a multistage process that involves scrubbing data, estimating parameters, reviewing fit, and exploring multivariate relationships is developed. The process reveals attribute misunderstandings, considers candidate training targets for subsequent training sessions, and explores data to better understand product properties.

Careless responders have a large impact on a study by causing issues such as Type II errors (failing to reject a false null hypothesis), which then waste researchers’ time and money. Research on careless responding has focused primarily on detecting and removing careless responders rather than on reducing careless responding before data collection begins. The purpose of the present study was to test the use of honor codes with or without the presentation of a picture of watchful eyes to increase self-awareness thereby reducing careless responding. Participants (N = 305) were randomly assigned to one of five honor code conditions (control condition, read-only condition, type condition, read-and-eyes condition, or type-and-eyes condition) and then completed a number of personality measures. Participant’s responses were screened for careless responding. I found that when participant’s only read an honor code without a picture of watchful eyes on the screen, they were significantly more likely to engage in careless responding than were people in the control condition. There was no significant difference in careless responding in the control condition compared to the other three conditions (type condition, read-and-eyes condition, or type-and-eyes condition). This finding indicates that participants being presented only with an honor code, and no other cues of moral behavior, might lead to psychological reactance (i.e., participants perceiving their freedom or control was threatened). This psychological reactance was likely due to participants feeling coerced to respond honestly by the honor code rather than participants responding honestly on their own accord. Additionally, there were no differences on participant’s responses across conditions on a number of personality measures that may be sensitive to increased self-awareness. A number of exploratory analyses were also conducted to further examine psychological reactance. The present study provides answers and possible directions to a number of useful questions that could improve data quality and reduce potential negative effects of honor codes under certain circumstances.

Survey data quality is influenced by the care and attention that respondents take in answering questions. Careless and inattentive (CI) responding is a confound in survey data that can distort findings and lead to incorrect conclusions. This quantitative study explored CI responding in job analysis studies supporting occupational certification programs and its relationship to survey features, data quality measures, and test content validity. Satisficing theory served as the framework, and secondary analysis of 3 job analysis surveys was undertaken. Results indicated that 9-33% of respondents engaged in CI responding, with the rate differing by CI index used (Mahalanobis distance, long string analysis, or person-total correlation) and by occupation. Each index detected a distinct pattern of carelessness, supporting the use of multiple indices. The indices performed best detecting carelessness in frequency ratings and may not be useful for all job analysis rating scales. Partial support was found for relationships between carelessness and survey features. CI responding had a minimal impact on mean ratings, correlations, and interrater reliability, and had no impact on certification test content outlines. By providing guidance and caution on the use of CI response detection methods with job analysis survey data, this study produced two potential avenues for social change. For practitioners conducting occupational job analyses, the use of CI detection methods can enhance the validity of data used to make certification decisions. For researchers, follow-up studies can yield a more nuanced understanding of the most appropriate use of these methods in the job analysis context.

The business data could be considered as the raw material for decision-making process, for the development of corporate strategies and overall running of the business. Therefore, adequate attention should be paid to quality of the data. The main goal of the diploma thesis is elaboration of a specific framework for data quality assurance, which combines three theoretical concepts: time series analysis, data screening and data profiling -- business-specific data profiles are monitored by data screening during the data warehouse ETL (extract, transform and load) process and results are afterwards compared with the values predicted by time series analysis. Achievement of this goal is based on the analysis of "data quality" in literature, exact problem definition and selection of appropriate means for its solution. Moreover, the thesis is analysing alternative solutions available on the market and comparing their functionality with the functionality of own framework, as well.

No description available.