• Proceedings of the Korean Statistical Society Conference
• /
• 한국통계학회 2003년도 추계 학술발표회 논문집
• /
• pp.123-130
• /
• 2003

We define a broad class of rotation designs whose monthly sample is balanced in interview time, level of recall, and rotation group, and whose rotation scheme is time-invariant. The necessary and sufficient conditions are obtained for such designs. Using these conditions, we derive a minimum variance unbiased generalized composite estimator (MVUGCE). To examine the existence of time-in-sample bias and recall bias, we also propose unbiased estimators and their variances. Numerical examples investigate the impacts of design gap, non-sampling error sources, and two types of correlations on the variance of MVUGCE.

• Journal of The Korean Society of Integrative Medicine
• /
• 제10권4호
• /
• pp.71-81
• /
• 2022

Purpose : This study aimed to present evidence by analyzing the characteristics and effectiveness of group art therapy interventions through an examination of domestic studies on group art therapy for older people with dementia. Methods : The database used DBpia, Riss, and Google Scholar, and the research period was from 2016 to November 2021. For the selected studies, the level of evidence was analyzed, bias evaluation was performed, and patient, intervention, comparison, and outcome were analyzed. For the evaluation of bias, the risk of bias assessment tool for non-randomized study (RoBANS) and Cochrane's risk of bias (RoB) were used. Results : As for the level of evidence of the included studies, level I consisted of five studies, and levels II and III each had one article. As a result of the bias evaluation of five studies through RoB, a "low risk of bias" was found for incomplete result data, selective result reporting, and others, except for four unclear evaluation areas. The "low risk of bias" ratio was 0~25 % in the evaluation of bias in two studies through RoBANS. For the evaluation tool, cognitive evaluation tool was used the most while mini-mental state examination-Korea was used the most frequently. For the intervention method, the most frequently used was group art therapy that employed recall in three studies, while collage, Korean painting, use of paper media, and procedural memory were used in each of the other studies. Each intervention was found to be significantly effective overall. Conclusion : This study provided clinical evidence by systematically reporting research on group art therapy for older people with dementia. In the future, it is necessary to check the effect of group art therapy on various areas other than cognition for older people with dementia. Moreover, the study should be conducted with the risk of bias sufficiently taken into consideration.

• Journal of the Korean Statistical Society
• /
• 제32권3호
• /
• pp.245-259
• /
• 2003

The 2-way balanced one-level rotation design has been discussed (Park et al., 2001), where the 2-way balancing is done on interview time in monthly sample and rotation group. We extend it to 3-way balanced multi-level design to obtain more information of the same sample unit for one or more previous months. The 3-way balancing is accomplished not only on interview time in monthly sample and rotation group but also on recall time as well. The 3-way balancing eliminates or reduces any bias arising from unbalanced interview time, rotation group and recall time, and all rotation groups are equally represented in the monthly sample. We present the rule and rotation algorithm which guarantee the 3-way balancing. In particular, we specify the necessary and sufficient condition for the 3-way balanced multi-level rotation design.

• Journal of Nutrition and Health
• /
• 제32권8호
• /
• pp.957-966
• /
• 1999

Underreporting of dietary intake is common and might distort analysis and interpretation of dietary surveys. This study was designed to investigate the degree of underreporting and characterastics of under-reporting group in Korean college female students. Dietary survey of 1-day 24-hour recall method was conducted on 379 college students in Seoul and Chonan areas. Physical activity and life style were aquired from questionnnaires. Underreporting was defined as energy intake(EI) lower than 0.9BMR(based metabolic rate), since EI<0.98BMR is statistically judged as bias in 1-day 24 hour recall. BMR was calculated from Schofield's equation. Proportion of underreporting was 18.7% and it's not so different from one of American or European women. Intake of nutrients except vitamin A by underreporting group was lower than other groups(p<0.001). Proportions of subjects with nutrient intake level less than 75% of Korean RDA were more than 80% in protein, Ca, Fe, vitamin A, riboflavin, niacin, zinc as well as energy. Dietary quality of underreporting group was also worse than other groups. Proportion of subjects less than 3 food groups among 5 food group was higher in underreporting group. The number of foods eaten by underreporting group were also less than those of other groups. BMI and body weight were the largest in underreporting group(p<0.05) and the trial of weight reduction was shown higher trend(p<0.01). Different in PAC and other characteristics between underreporting group and other group were not significant. Not only dietary quantity but also dietary quality were worse in the underreporting group. Furthermore underreporting in college female students seemed to be affected by body weight and concern for weight reduction.

• Nutrition Research and Practice
• /
• 제16권5호
• /
• pp.646-657
• /
• 2022

BACKGROUND/OBJECTIVES: The doubly labeled water (DLW) method is the gold standard for estimating total energy expenditure (TEE) and is also useful for verifying the validities of dietary evaluation tools. In this study, we compared the accuracy of total energy intakes (TEI) estimated by the 24-h diet recall method with TEE obtained using the doubly labeled water method. SUBJECTS/METHODS: This study involved 71 subjects aged 20-49 yrs. Over a 14-day period, three 24-h diet recalls per subject (2 weekdays and 1 weekend day) were used to estimate energy intakes, while TEE was measured using the DLW method. The paired t-test was used to determine the significance of differences between TEI and TEE results, and the accuracy of the 24-h recall method was determined by accuracy predictions percentage, root mean square error, and bias. RESULTS: Average study subject age was 33.4 ± 8.6 yrs. The association between TEI and TEE was positive and significant (r = 0.463, P < 0.001), and the difference between TEI (2,084.3 ± 684.2 kcal/day) and TEE (2,401.7 ± 480.3 kcal/day) was also significant (P < 0.001). In all study subjects, mean TEI was 12.0% (307.5 ± 629.3 kcal/day) less than mean TEE, and 12.2% (349.4 ± 632.5 kcal/day) less in men and 11.8% (266.7 ± 632.5 kcal/day) less in women. Rates of TEI underprediction for all study subjects, men, and women, were 60.5%, 51.4%, and 66.7%, respectively. CONCLUSIONS: This study shows that 24-h diet recall underreports energy intakes. More research is needed to corroborate our findings and evaluate the accuracy of 24-h recall with respect to additional demographics.

• The Korean Journal of Applied Statistics
• /
• 제17권1호
• /
• pp.61-73
• /
• 2004

Rotation sampling designs may be classified into two categories. The first type uses the same sample unit for the entire life of the survey. The second type uses the sample unit only for a fixed number of times. In both type of designs, the entire sample is partitioned into a finite number(=G) of rotation groups. This paper is generalization of the first type designs. Since the generalized design can be identified by only G rotation groups and recall level 1, we denote this rotation system as l/G rotation design. Under l/G rotation design, variance and mean squared error (MSE) of generalized composite estimator are derived, incorporating two type of biases and exponentially decaying correlation pattern. Compromising MSE's of some selected l/G designs, we investigate design efficiency, design gap effect, ans the effects of correlation and bias.

• 한국노년학
• /
• 제38권1호
• /
• pp.83-102
• /
• 2018

According to socioemotional selectivity theory, if people perceive their time left in life as expanded, they have a future-oriented goal of life, but if perceive as limited the goal of life is changed into the pursuit of present emotional satisfaction. Thus, if we perceive our time left as getting limited as we get older, we pay more attention to the positive stimuli than the negative ones and remember more the positive stimuli in order to maintain the current emotional state as positive. This is known as the positivity effect. This study examined whether the positivity effect is caused by a limited future time perspective. The participants were presented with scenarios for hypothetical situations in which the future time was expanded or limited, and were encouraged to immerse in the virtual situation by talking about what they would like to do and whom they wanted to spend time with. Then the participants were presented with 48 positive, negative, and neutral emotional pictures and were asked to recall after 10 minutes delay. 75 university students and 65 elderly participated in the study. In the control condition where the future time perspective was not manipulated, the elderly showed the positivity effect but the youth showed the bias toward negative pictures. The elderly in the expanded time condition recalled positive pictures less and negative pictures more than the elderly in the control condition. On the other hand, the youth in the limited time condition recalled less the negative pictures than the youth in the control condition. These results demonstrated that the elderly did not show the positive bias when the future time perspective was expanded, and that the youth showed the positive bias when the future time perspective was limited. These results show that the positivity effect is related with the limited future time perspective.

• Journal of Environmental Health Sciences
• /
• 제45권1호
• /
• pp.71-81
• /
• 2019

Objectives: The lung injuries by exposure to the humidifier disinfectants (HDs) were reported in 2011, Korea. For the HD victims, environmental exposure level and clinical diagnosis were conducted to determine the levels of damage by HDs. Methods: The exposure assessment to the HDs from 1st to 4th questionnaire surveys were carried out for 5,245 victims. And the affecting factors of exposure levels were analyzed by characterizing exposure and demographic information. By using of exposure concentration and cumulative time, exposure levels were classified and compared by percentage of clinical diagnosis classes. The high exposure and low clinical diagnosis rating groups, and low exposure and high clinical diagnosis rating groups were analyzed to overcome the limitation of past exposure assessment such as recall bias. Results: Among the all applicants damaged by the humidifier disinfectants, survivors were 4,028 and the dead were 1,217. And male and female were 2,675, and 2,547, respectively. In case of occurrence age of lung disease, under 10 years was majority age group (1,536) and followed by thirties (917). Pregnant women and fetuses were 339 and 439, respectively. And the damages by exposure to the HDs were concentrated on these susceptible populations in groups with low exposure and high clinical diagnosis rating. On the other hand, the groups classified by high exposure and low clinical diagnosis rating were shown different characterization. Conclusions: The questionnaire survey on past exposure may be uncertain due to recall bias. However, the relationship between classified exposure levels and clinical diagnosis ratings might be shown positive correlation if the exposure assessment errors were analyzed and controlled.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• 제21권12호
• /
• pp.320-330
• /
• 2020

Fire is difficult to achieve good performance in image detection using deep learning because of its high irregularity. In particular, there is little data on fire detection in underground utility facilities, which have poor light conditions and many objects similar to fire. These make fire detection challenging and cause low performance of deep learning models. Therefore, this study proposed a fire detection model using deep learning and estimated the performance of the model. The proposed model was designed using a combination of a basic convolutional neural network, Inception block of GoogleNet, and Skip connection of ResNet to optimize the deep learning model for fire detection under underground utility facilities. In addition, a training technique for the model was proposed. To examine the effectiveness of the method, the trained model was applied to fire images, which included fire and non-fire (which can be misunderstood as a fire) objects under the underground facilities or similar conditions, and results were analyzed. Metrics, such as precision and recall from deep learning models of other studies, were compared with those of the proposed model to estimate the model performance qualitatively. The results showed that the proposed model has high precision and recall for fire detection under low light intensity and both low erroneous and missing detection capabilities for things similar to fire.

• Journal of the Korea Society of Computer and Information
• /
• 제28권2호
• /
• pp.9-17
• /
• 2023

In this paper, we propose a one-class vibration anomaly detection system for bearing defect diagnosis. In order to reduce the economic and time loss caused by bearing failure, an accurate defect diagnosis system is essential, and deep learning-based defect diagnosis systems are widely studied to solve the problem. However, it is difficult to obtain abnormal data in the actual data collection environment for deep learning learning, which causes data bias. Therefore, a one-class classification method using only normal data is used. As a general method, the characteristics of vibration data are extracted by learning the compression and restoration process through AutoEncoder. Anomaly detection is performed by learning a one-class classifier with the extracted features. However, this method cannot efficiently extract the characteristics of the vibration data because it does not consider the frequency characteristics of the vibration data. To solve this problem, we propose an AutoEncoder model that considers the frequency characteristics of vibration data. As for classification performance, accuracy 0.910, precision 1.0, recall 0.820, and f1-score 0.901 were obtained. The network design considering the vibration characteristics confirmed better performance than existing methods.