China
Africa
Explore chapters and articles related to this topic
Usability for Engaged Users
Published in Julie A. Jacko, The Human–Computer Interaction Handbook, 2012
In contrast to cross-sectional research, longitudinal research follows a given sample over time so that the evolution of each individual can be tracked. This approach has many challenges of its own. The demands of participation that will extend over a longer period may make it harder to find willing participants, and thus, introduce a selection bias. In any sample followed over time, there will be attrition, with people dropping out for various reasons. Some of these reasons may be due to highly individual changing circumstances that make participation in the study difficult and are, therefore, in a sense “accidental” or random. But some of the reasons may be correlated with user experience issues. In many real life situations, people are free to abandon a tool to return to their previous one or a new one based on their experiences. As mentioned earlier, this may be less true in some business situations, but even in business settings people often have some freedom about how exclusively they rely on a tool and how many of its capabilities they explore and adopt.
Data Statistics and Analytics
Published in Paresh Chra Deka, A Primer on Machine Learning Applications in Civil Engineering, 2019
Time series analysis is a statistical technique that represents time-series data or a trend analysis. This will represent the raw data in a graphical view. This method is used to analyze time-series data in order to extract meaningful information. Time is represented by the X-axis and the data series observations are represented by the Y-axis. ‘Time series’ means that the data is from a series of particular time periods or intervals. Time series, cross sectional, and pooled data are the types of data considered for the analysis. In time-series data, the data is from observations of a variable at different periods of time. Cross-sectional data represents multiple variables collected at the same point in time. Pooled data is a combination of time series and cross-sectional data. Time series forecasting is applied to predict future values based on values observed in the past. An interrupted time series is the analysis of interventions on a single time series. They are different from spatial data analysis where the observations typically relate to geographical locations. A stochastic model in a time series is more accurate when the observations are closer than when they are further apart. A time-series analysis can be applied to real-valued, continuous, and discrete numeric data. There are two solution techniques for time series: frequency domain and time-domain methods. Spectral and wavelet methods are the methods of solution followed in a frequency domain technique, whereas autocorrelation and cross-correlation are applied to the domain method. Figure 7.2 shows time series analysis with random data and a best-fit line.
Reorganization and downsizing in the petroleum sector
Published in Stein Haugen, Anne Barros, Coen van Gulijk, Trond Kongsvik, Jan Erik Vinnem, Safety and Reliability – Safe Societies in a Changing World, 2018
L.I.V. Bergh, R. Høydal, J.E. Tharaldsen, C. Aagestad, T. Sterud
This study is cross-sectional and provide a snapshot of a particular group at a given time. In research, cross-sectional studies are used in order to determine prevalence. Cross sectional studies are the best way to determine prevalence and are useful at identifying associations that can then be more rigorously studied using a cohort study or randomized controlled study (Mann 2003). As such, this may limit the ability to draw firm conclusions about relationships observed between exposure (for example, downsizing and reorganization) and outcomes (for example, occupational injuries), and the extent to which exposure precedes outcome (for example, occupational injuries) in time. Given that the data was cross sectional, we cannot draw conclusions about causal relationships. Thus, there is a possibility that those who have been injured during the reorganization are somewhat more “negative” when they answer the survey or that the changes have more and more affected groups that initially reported poorer psychosocial work environment and safety. However, it is important to note that research over at least the last decade, including longitudinal studies, has shown that psychosocial hazards can have a negative impact on health and safety (Leka & Jain 2010, Mackey, Palverman Saul et al. 2012). Another important limitation is that those who are absent from work due to sickness, health complaints or injury during the survey are not included in the data. As such, studying potential consequences of downsizing or termination of employment is difficult, because those who has been terminated are not included in the data. In other words, those remaining after the downsizing processes are those included in the data.
Optimize the online shopping title of men’s plain-color shirts in e-commerce based on Kansei Engineering
Published in Journal of Global Fashion Marketing, 2023
Simple random sampling is probability sampling, that is, each member of the population has an equal chance to be selected (Creswell, 2017). A 5-point Likert scale questionnaire is a common instrument used in simple random sampling. The online questionnaire is a safe and effective form of the questionnaire under special circumstances. Longitudinal and cross-sectional studies are two ways of collection for sampling. In longitudinal studies, data are collected repeatedly from the same sample at different times; in cross-sectional studies, data are collected from the population at specific time points. Therefore, this study will use a 5-point Likert scale web questionnaire, cross-sectional study to conduct a simple random sampling focus on consumers aged 20-35 who have purchased men’s plain-color shirts during online shopping in the past year online.
Identifying childhood movement profiles and comparing differences in mathematical skills between clusters: A latent profile analysis
Published in Journal of Sports Sciences, 2021
Timo Jaakkola, Airi Hakkarainen, Arto Gråsten, Elina Sipinen, Anssi Vanhala, Mikko Huhtiniemi, Anu Laine, Kasper Salin, Pirjo Aunio
This study has few limitations. First, we were not able to measure socioeconomic status of students’ families. Secondly, cross-sectional data does not allow us to draw causal conclusions. Thirdy, it should be recognised that “skilled movers” and “expert movers” profiles were rather small, which weakens the statistical power of the analyses. Lastly, one of the limitations is that we only measured maths performance. Future studies should also investigate if other academic domains (e.g., reading) have associations with students’ motor performance. In future studies, it would be beneficial to add open problems with multiple correct answers to ProbSol instrument to enhance students’ creativity. Additionally, previous studies have demonstrated that PA interventions have also contributed to academic performance34, including mathematical skills (Ericsson and Karlsson, 2014). In the future, these intervention studies should be targeted especially for students who have weak HRF and MC and investigate whether increases in PA engagement, and subsequently, physical performance have a positive association with BasicMath and ProbSol skills.
Countervailing Risk Management Through Knowledge Transfer
Published in Engineering Management Journal, 2020
Jeffery Temple, Rafael E. Landaeta
In a field study, data can be collected using different tools; a survey or questionnaire is an effective tool to collect data about several variables from a relatively large number of potential respondents. Bowen (1995) suggests that surveys provide an opportunity to study a large number of groups providing the strength of high external validity, assuming the data samples include multiple organizations and settings. “Surveys include cross-sectional and longitudinal studies using questionnaires or structure interviews for data collection, with the intent of generalizing from a sample to a population” (Creswell, 2003, p. 14). Cross-sectional studies are performed at a point in time and are a snapshot as opposed to longitudinal studies that make observations over time. For this study, the intent was to perform a cross-sectional (survey) study to collect data, with the population sample (i.e., unit of analysis) being projects in a large organization within the Department of the Navy that have performed continuous improvement Lean Thinking implementation projects.