Hello,
I start working on the data set that will be a part of my thesis writing. The sample size of data is 4700 that have been collected in two waves/times.
My first data source was a cross-sectional set and I utilized SEM in AMOS. Although I applied Hayes (2013) for one of my hypotheses related to serial mediation.
Now I would like to ask it is possible to utilize serial mediation for the current longitudinal data set. I want to test the moderation role of the independent variable in the model as well. The whole of the model consists of one independent variable, two mediators, and three outcomes, altogether six latent constructs. Some demographic variables may use as control variables. It is highly appreciated for guiding me with the below-mentioned questions. I don’t know which source can help me to get familiar with analyzing longitudinal data in AMOS.
Some of my questions are related to:
What kind of data screening are essential for panel studies;
How do I run CFA for the longitudinal study;
How do I consider control variables in the model;
What approach do you suggest for serial mediation and interaction in the longitudinal set?
Sharing any references, papers, or information will help me to know how to work on the current data set.
Thanks
Hi, Shaghayegh Ghaffaripour, depending on your needs, I recommend you read the following papers. Also, I have uploaded a book --Applied Longitudinal Data Analysis, I hope it will help you.
Lock, D., Funk, D. C., Doyle, J. P., & McDonald, H. (2014). Examining the longitudinal structure, stability, and dimensional interrelationships of team identification. Journal of Sport Management, 28(2), 119-135.
Fitzmaurice, G. M., & Ravichandran, C. (2008). A primer in longitudinal data analysis. Circulation, 118(19), 2005-2010.
O'Laughlin, K. D., Martin, M. J., & Ferrer, E. (2018). Cross-sectional analysis of longitudinal mediation processes. Multivariate Behavioral Research, 53(3), 375-402.
MacKinnon, D. P., & Fairchild, A. J. (2009). Current directions in mediation analysis. Current directions in psychological science, 18(1), 16-20.
@Ghaffaripour @Wang has given you a really good piece of advice. Having look at your query, Cross Lagged Panel Model seems appropriate. Refer to Ferrer et al. (2018; Number 4th on the list of suggested readings by @Wang). It's really a good paper that talks about various ways of analysing longitudinal mediation model. Wishing you best.
Wang Zheng
Hi,
Your recommended materials have clearly defined the method sections. Thanks for providing me with these effective resources.
Imran Anwar
Sure, I will focus on Cross Lagged Panel Model ,the approach available in the material Wang Zheng has already shared.
Thanks
Dear Shaghayegh Ghaffaripour would you like to share your time-lagged data collection procedure please? More specifically, how many variables you measured in how many time frames as it will be important for a cross lagged panel analysis.
Taking this opportunity I want to ask my dear colleagues Wang Zheng and Imran Anwar is there any difference between a Time-Lagged study and a Longitudinal study? Thank you.
Dear @Zia Aslam Cross Lagged Panel (CLPM) Model is nothing but a way of analysing repeated measures based longitudinal data with two or more time frames. Refer to below attachment. It's a very good paper for longitudinal mediation models.
Dear
Muhammad Zia Aslam,
The data collection of T1 was started in the autumn of 2020 and it repeated with the same individuals in Mach 2021.
The study is consisted of 14 constructs, overall 68 items, with some demographic questions.
Fix items of T1 and T2 are 47 and 11 items of three constructs measured only in T1 and 10 items of three constructs were added in T2.
Could I consider this type of data as a longitudinal study?
Thanks
I do not see two time points as longitudinal data , rather pre and post. That is I would use the pre values as a covariate. That should make things easier.
@ Kelvyn Jones,
Thank you for offering your valuable advice.
I would recommend reading the following article:
- Christophe Genolini · Bruno Falissard (2010) KmL: k-means for longitudinal data, Comput Stat 25:317–328. DOI 10.1007/s00180-009-0178-4
Dear Shaghayegh Ghaffaripour it is tricky to use terminology in a time lagged study, as most of the researchers do it just to avoid Common Method Bias and not for cross-lagged longitudinal interpretations. I will suggest you to read maximum quality published literature based on time lagged (Longitudinal) methods. It will help you craft your methods section in a more authentic way. Try to read Prof. Hannes Zacher's work. Tq.
Some of the recent relevant articles are given below.
Griep, Y., Lukic, A., Kraak, J. M., Bohle, S. A. L., Jiang, L., Vander Elst, T., et al. (2021). The chicken or the egg: The reciprocal relationship between job insecurity and mental health complaints. Journal of Business Research, 126, 170-186. https://doi.org/10.1016/j.jbusres.2020.12.045
Koziel, R. J., Friedrich, J. C., Rudolph, C. W., & Zacher, H. (2021). Age-differentiated leadership and healthy aging at work: Evidence from the early stages of the covid-19 pandemic. International Journal of Environmental Research and Public Health, 18(23), 12509. https://doi.org/10.3390/ijerph182312509
Dear Muhammad Zia Aslam,
Thanks for the helpful comments and for sharing recently published papers.
- Badges
- Science topic