Observational Studies
Abhaya Indrayan in Research Methods for Medical Graduates, 2019
The primary objective of a longitudinal study is to track the trend over time, generally of a quantitative measurement. This means that the time points are an important consideration. For example, a pharmacokinetic study that evaluates peak concentration of a drug and time to reach peak would require a longitudinal study because observations at several time points are needed for this kind of study. Similarly, a study on growth of children would need a longitudinal study to track their trajectory. In both these setups the outcome is quantitative, but that is not a prerequisite for a study to be longitudinal. Time-invariant risk factors such as gender and family history may be measured only at baseline and other associated risk factors may be measured repeatedly over time in a longitudinal study.
Longitudinal data
Pat Dugard, John Todman, Harry Staines in Approaching Multivariate Analysis, 2010
There are many advantages of using a longitudinal study. If each person is able to try more than one treatment or condition then they can act as their own control, and since variation within a participant is typically less than that between participants we may achieve greater sensitivity or power. Also, for a fixed number of participants, we obtain more information if we can record measures on several occasions. But most important, if we want to find out about a process for individuals, then a record of their progress is essential. If we want to know about growth or decline or progress on a treatment or training program, then a record that follows each individual over a period of time will be much more informative than data on different individuals who are measured at different stages of the process.
Introduction to the Analysis of Longitudinal Data
Lesa Hoffman in Longitudinal Analysis, 2015
The primary benefit of a longitudinal study is its capacity to inform about within-person relationships (and not just between-person relationships, as in cross-sectional studies). But another important benefit is that longitudinal studies provide the opportunity to test hypotheses at multiple levels of analysis simultaneously (see Hofer & Sliwinski, 2006). That is, the models in this text will allow us to examine both between-person and within-person relationships in the same variables at the same time. In fact, much of this text will emphasize the need to distinguish between-person from within-person relationships (as well as relationships at other levels when applicable), both in terms of specifying longitudinal research questions and in examining these relationships with longitudinal models. Although human development, psychology, and other fields are replete with theoretical explanations of human behavior, it can often be a challenge to identify theoretical implications at both the between-person and within-person levels of analysis.
Job satisfaction and organizational citizenship behaviour amongst health professionals: The mediating role of work engagement
Published in International Journal of Healthcare Management, 2021
Lee-Peng Ng, Yuen-Onn Choong, Lok-Sin Kuar, Chun-Eng Tan, Sok-Yee Teoh
There are several limitations of the study need to be addressed. First, the cross-sectional approach of this study is unable to provide a strict causal conclusion. Cross-sectional data is only capable in revealing the net effect of predictor variable towards a particular criterion variable at a specific point of time [49]. Hence, longitudinal study can be used in the future study. Second, it is possible that common method bias may exist in this study as the data are collected from a single source at one time, thus multiple-source of data collection (e.g. supervisor and peers) can be done in the future. Next, the current study only covered registered nurses and doctors from two hospitals which likely to limit the generalization of the data. Future research can cover a larger number of samples across different hospitals. For data collection, the response rate of the study is at 43% which mean that most of the health professionals have no time to participate in this survey. Apart from this, the majority of the respondents are aged between 21 and 30 as a result of convenient sampling technique used in the study. In order to generate better result, future researchers can adopt quota sampling technique to ensure an equal sample size from each age group. Lastly, global measure of JS is used in this study in view that the interest of the researcher is to understand the overall feelings of employees about their job [50], further study can be conducted by evaluating the multiple-facet of JS on higher-order construct of WE and OCB as well as their dimensions.
A Bayesian conditional model for bivariate mixed ordinal and skew continuous longitudinal responses using quantile regression
Published in Journal of Applied Statistics, 2018
S. Ghasemzadeh, M. Ganjali, T. Baghfalaki
Longitudinal study has a wide range of applications in medicine, epidemiology and economics. It is a study wherein variables of interest are repeatedly measured for independent subjects over time and the measurements are correlated for the same subject. In the last decades, the use of the QR model in longitudinal data analysis has been increased. For example, Lipsitz et al. [20] described the use of weighted estimating equations in the QR model for CD4 cell count data with dropouts. Koenker [16] proposed the 14] presented a likelihood-based approach to make inference of parameters by using ALD for the errors in a longitudinal data analysis. Yuan and Yin [34] employed a QR model for a longitudinal data with non-ignorable intermittent missing data. Luo et al. [21] discussed a fully Bayesian QR model using the MCMC method for longitudinal data. Aghamohammadi and Mohammadi [1] proposed penalized Bayesian QR model with random effects for longitudinal data. Farcomeni and Viviani [9] proposed a joint model for a continuous response and time-to-event outcome based on connection between ALD and QR model.
Exploring the relationship between social support and mental health status among lymphoma survivors: Does patient-centered communication really matter? A brief report
Published in Journal of Psychosocial Oncology, 2023
Nicole Caviness-Ashe, Sheryl Zimmerman, Lolita Chappel-Aiken, Elijah O. Onsomu, Ashley Leak Bryant, Sophia K. Smith
The present study contained several limitations and strengths. The sample stemmed from a study at two southern academic medical centers, which may contribute to sampling bias; however, the sample closely resembles the population of NHL survivors. Second, some participants in the 2005 baseline survey were unable to participate (i.e., too ill, deceased) in the 2010 follow-up survey. Because their responses could not be included, this study’s data may be skewed, reflecting data from healthier survivors. Nonetheless, the response rate (83%) was high. Third, the cross-sectional study design does not afford the opportunity to examine how PCC, social support, and mental health outcomes have changed over time. A longitudinal study of the variables is needed to examine possible changes over time. Fourth, the NHL representation was 87% White and does not fully capture PCC in racial/ethnic minorities.
Related Knowledge Centers
- Clinical Psychology
- Cohort Effect
- Cohort Study
- Developmental Psychology
- Observational Study
- Prospective Cohort Study
- Retrospective Cohort Study
- Randomized Experiment
- Opinion Poll
- Cross-Sectional Study