• Survey Research
• /
• v.9 no.2
• /
• pp.51-83
• /
• 2008

While online surveys are considered to be the viable alternative to face-to-face and telephone surveys, many clients express a strong concern about the possibility of unreliable responses. Given this concern, survey experts hypothesize that amounts of time on responding can represent the extent to which respondents fill out the questions with sincerity. This study examines this hypothesis using data from an online survey sample of 1.052. Results strongly suggest that while "long time" informants are reliable, "short time" informants are not. Those in the short time group spend only 2.4 seconds per question and show very different patterns of responses to 49 Likert -typed questions. Logistic regression results imply that these unreliable respondents are interested in monetary gifts and maybe as a result, participate in surveys much more frequently. In conclusion, we provide some implications of these results to the quality of online surveys.

• The KIPS Transactions:PartD
• /
• v.12D no.4 s.100
• /
• pp.657-662
• /
• 2005

The Non-Sampling Error can happen any time by means of the intended or unintended error by the interviewer or respondent, but it is very difficult to find the error in survey database because it can hardly be computed mathematically and systematically. Until now, we have found it accidentally through the simple relation between the items or through the inspection from the random field. Therefore we introduced an heuristic methodology that can detect the interviewer's error by statistical decision-making or data mining techniques with a case study. It will be helpful so as to improve the statistical duality and provide efficient field management for the supervisor.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2009.03a
• /
• pp.146-153
• /
• 2009

One dimensional site response analysis is widely used in prediction of the ground motion that is induced by earthquake. Equivalent linear analysis is the most widely used method due to its simplicity and ease of use. However, the equivalent linear method has been known to be unreliable since it approximates the nonlinear soil behavior within the linear framework. To consider the nonlinearity of the ground at frequency domain, frequency dependent algorithms that can simulate shear strain - frequency dependency have been proposed. In this study, the results of the modified equivalent linear analysis are compared to evaluate the degree of improvement and the applicability of the modified algorithms. Results show the novel smoothed curve that is proposed by this study indicates the most stable prediction and can enhance the accuracy of the prediction.

• The Journal of the Korea Contents Association
• /
• v.17 no.11
• /
• pp.60-69
• /
• 2017

Recently, with the rapid development of internet and smart phones, social network services that can create and share various information through relationships among users have been actively used. Especially as the amount of information becomes enormous and unreliable information increases, expert recommendation that can offer necessary information to users have been studied. In this paper, we propose an expert recommendation scheme considering users' interests, human relations, and response quality. The users' interests are evaluated by analyzing their past activities in social network. The human relations are evaluated by extracting the users who have the same interesting fields. The response quality is evaluated by considering the user's response speed and response contents. The proposed scheme determines the user's expert score by combining the users' interests, the human relations, and the response quality. Finally, we recommend proper experts by matching queries and expert groups. It is shown through various performance evaluations that the proposed scheme outperforms the existing schemes.

• International Journal of Precision Engineering and Manufacturing
• /
• v.5 no.4
• /
• pp.61-68
• /
• 2004

Necessity to address engineering system uncertainties in design processes has long been acknowledged. To obtain quality of product, a safety factor is traditionally used by many design engineers due to its easy of use and comprehension. However, the safety factor approach often yields either conservative or unreliable designs, since it ignores the type of probability distribution and the mechanism of uncertainty propagation from the input to the output. For a consistent reliability-based design, two fundamental issues must be investigated thoroughly. First, the design-decision process that clearly identifies a mechanism of uncertainty propagation under system uncertainties needs to be developed, which must be an efficient and accurate process. To identify the mechanism more effectively, an adaptive probability analysis is proposed by adaptively setting probability levels through a posteriori error estimation. The second is to develop the design process that not only yields a high quality design but also a cost-effective optimum design from manufacturing point of view. As a result, a response surface methodology is specially developed for RBDO, thus enhancing numerical challenges of efficiency and complicatedness. Side crashworthiness application is used to demonstrate the integrated design process for product and manufacturing process design.

• Asia-Pacific Journal of Business
• /
• v.13 no.4
• /
• pp.109-121
• /
• 2022

Purpose - This study investigates how earnings volatility influences current stock price informativeness about expectations of future earnings. Design/methodology/approach - I adopt the FERC model developed by Collins et al. (1994) and modified by Lundholm and Myers (2002) to investigate the connection between earnings volatility and future earnings reflected in current returns. I define five-year rolling standard deviations of earnings and components as earnings volatility measures and the degree of deviation of earnings from cash flows over the same five-year, which is developed by Jayaraman (2008). Finding - My results show that earnings volatility delays current stock price response to future operation expectations. They also verify that as earnings are more divergent from cash flows, current returns are less timely incorporating value-relevant future operation. Research implications or Originality This study shows that when volatile earnings deliver obscure and unreliable information about future operation expectations, they cause the market to be conflicting in understandings their implications and make it difficult in attaining correct future cashflow estimates.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.32 no.6
• /
• pp.648-660
• /
• 2012

Reliable response measurements are extremely important for proper bridge health monitoring but incomplete and unreliable data may be acquired due to sensor problems and environmental effects. In the case of a sensor malfunction, parts of the measured data can be missing so that the structural health condition cannot be monitored reliably. This means that the dynamic characteristics of natural frequencies can change as if the structure is damaged due to environmental effects, such as temperature variations. To overcome these problems, this paper proposes a systematic procedure of data analysis to recover missing data and eliminate the environmental effects from the measured data. It also proposes a health index calculated statistically using revised data to evaluate the health condition of a bridge. The proposed method was examined using numerically simulated data with a truss structure and then applied to a set of field data measured from a cable-stayed bridge.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.2
• /
• pp.619-634
• /
• 2019

In the field of traceability systems, researchers focus on applications in the agricultural food traceability and scanning commodities. The purposes of this paper, however, is to propose an efficient and reliable traceability system that can be applied to all kinds of commodities. Currently, most traceability systems store data in a central server, which is unreliable when the system is under attack or if the administrator tampers with the data for personal interests. Therefore, it is necessary to design a system that can eliminate these threats. In this paper, we propose a decentralized and non-reversible traceability system for storing commodity data. This system depends on blockchain technology, which organizes data in the form of chains without a central server. This chain-style storage mechanism can prevent malicious modifications. In addition, some strategies are adopted to reduce the storage pressure and response time when the system has stored all kinds of commodity data.

• Journal of Technologic Dentistry
• /
• v.41 no.4
• /
• pp.305-317
• /
• 2019

Purpose: The purpose of this study is to provide basic data for the improvement of dental care service for patient care. Methods: The survey period was from December 10, 2018 to June 10, 2019. The subjects were 180 patients who were treated with dental implant in 20 dental hospitals and clinics in Daejeon and Gyeonggi province. A total of 145 questionnaires were analyzed except for 35 copies which were unreliable or difficult to analyze. Results: The most important care method for prolonging the life time of dental implant prosthesis was the highest rate of correct brushing (39.4%), and a majority in the subjects used oral hygiene products (91.0%). With regard to discomfort after dental implant prosthesis, food particles between the teeth showed the highest response (49.7%), and discomfort with gum infections and bleeding and reduced mastication increased significantly with age (p<0.05), whereas feeling of irritation decreased significantly with age (p<0.05). Conclusion: This study is meaningful in that it approached the perception and discomfort of prosthesis from the patients' point of view. It needs to establish the basic data to improve dental treatment service for patient care.

• The Korean Journal of Applied Statistics
• /
• v.19 no.2
• /
• pp.331-338
• /
• 2006

Quantification method III is designed by C. Hayashi as visualizing technique for two-way cross-classified tables. Specially in Japan, its usefulness is timely proven in social and marketing surveys. In several instances, relatively large quantification scores are assigned to low-frequency categories. Thus, they lead to unreliable data interpretation. The aim of this study is to develop stable solution to overcome such traits of quantification method III. The solution is of shrinkage type induced by small perturbations and is applied to a multiple response data obtained in a Korean social survey.