This is an AQMEN (Data Science for Social Research) repository for Longitudinal Data Analysis for Social Scientists
Across the social sciences there is widespread agreement that longitudinal data provides powerful research resources. Many books and courses on longitudinal data analysis are oriented towards statistics and they often lack adequate coverage of the common practical challenges associated with undertaking βrealβ research. This workshop is specially designed for social scientists. It provides a brief and highly accessible introduction to quantitative longitudinal data analysis using Stata software
The research value of longitudinal data
Sources of longitudinal data
Analysing repeated cross-sectional data
Duration models
Panel data models
The workflow in longitudinal data analysis
Getting started using Stata
Practical exercises in longitudinal data analysis
The course will be supported by
Gayle, V. and Lambert, P. (2018) What is Quantitative Longitudinal Data Analysis? Bloomsbury Publishing.
Available at Amazon https://tinyurl.com/y8cbf7c2 or directly from Bloomsbury https://tinyurl.com/y9x5fltk
WARNING!
No research micro data will be distributed during this workshop.
The files required are available in this git repository.
Participants MAY NOT make copies of these files or distribute them.
The files have been specifically created by AQMEN for training and MUST not be used for social science research.
THE DATA ARE NOT REAL! DO NOT USE THEM FOR REAL ANALYSES!
We advise that participants galvanise their training experience and work with 'genuine' data files which they can dowload legally from the UK Data Service (http://ukdataservice.ac.uk/ - the old site is www.esds.ac.uk but you will be redirected).
At workshops where real survey datasets are used participants MUST sign a data access agreement.
Real administrative data are never used in training settings.
Files:
File Title | File Type | Description |
---|---|---|
longitudinal_lectures | Day 1 pdf of Powerpoint slides | |
stata_intro | Day 1 pdf of Powerpoint slides | |
workflow | Day 1 pdf of Powerpoint slides | |
workflow_paper | Gayle and Lambert (2017) | |
longitudinal | .do | Main instruction file for Day 2 |
longitudinal_text | .txt | Plain text (txt) version of the Stata .do file |
college_skills | .dta | Duration data example |
income_ft_male_panel_fake | .dta | Panel data example (fake BHPS data) |
psid_extract | .dta | Example from Cornwell and Rupert (1988) |
panel_greene | .dta | Example from Greene (1999) |
wemp | .dta | Panel data example (binary outcome) |
aindresp_example | .dta | Wave A of BHPS example data file (fake BHPS data) |
bindresp_example | .dta | Wave B of BHPS example data file (fake BHPS data) |
cindresp_example | .dta | Wave C of BHPS example data file (fake BHPS data) |
British Household Panel Study https://www.iser.essex.ac.uk/bhps/
Understanding Society - The UK Household Longitudinal Study https://www.understandingsociety.ac.uk/
CLOSER https://www.closer.ac.uk/
US Panel Study of Income Dynamics https://psidonline.isr.umich.edu/
The Cross National Equivalence Files https://cnef.ehe.osu.edu/
The Household, Income and Labour Dynamics in Australia (HILDA) https://melbourneinstitute.unimelb.edu.au/hilda
These links are to Amazon (https://www.amazon.co.uk) but there are a number of other reputable academic book sellers (e.g. https://blackwells.co.uk/bookshop/shops/).
Gayle, V. and Lambert, P., 2018. What is Quantitative Longitudinal Data Analysis?. Bloomsbury Publishing.
https://www.amazon.co.uk/Quantitative-Longitudinal-Analysis-Research-Methods/dp/1472515404/ref=sr_1_1?s=books&ie=UTF8&qid=1530726864&sr=1-1&keywords=vernon+gayle
My recent book on longitudinal data analysis (using Stata).
Kohler, U. and Kreuter, F., 2012. Data analysis using Stata. Stata press.
https://www.amazon.co.uk/Data-Analysis-Using-Stata-Third/dp/1597181102/ref=sr_1_1?s=books&ie=UTF8&qid=1530725888&sr=1-1&keywords=kohler+and+kreuter
This is a first class text book. It is clearly written and very comprehensive and has successfully been used as a core textbook on several courses that I have taught.
Mehmetoglu, M. and Jakobsen, T.G., 2016. Applied statistics using Stata: a guide for the social sciences. Sage.
https://www.amazon.co.uk/Applied-Statistics-Using-Stata-Sciences/dp/1473913233/ref=sr_1_2?s=books&ie=UTF8&qid=1530725649&sr=1-2&keywords=Stata
This is a first class text book. It is clearly written and very comprehensive
Gayle, V.J. and Lambert, P.S. (2017) The Workflow: A Practical Guide to Producing Accurate, Efficient, Transparent and Reproducible Social Survey Data Analysis. NCRM Working Paper. NCRM.
http://eprints.ncrm.ac.uk/4000/
A practical guide to the data analysis workflow.
Long, J.S. and Long, J.S., 2009. The workflow of data analysis using Stata. College Station, TX: Stata Press.
https://www.amazon.co.uk/Workflow-Data-Analysis-Using-Stata/dp/1597180475/ref=sr_1_1?s=books&ie=UTF8&qid=1530726163&sr=1-1&keywords=stata+workflow
A fantastic book. This is the 'bible' of good data analysis workflow practices.
https://stats.idre.ucla.edu/stata/
https://www.stata.com
https://www.stata.com/links/resources-for-learning-stata/
http://www.restore.ac.uk/Longitudinal/
https://www.stata.com/links/resources-for-learning-stata/
https://geocenter.github.io/StataTraining/portfolio/01_resource/
https://www.datasciencecentral.com/group/resources/forum/topics/stata-cheat-sheet
http://www-personal.umich.edu/~agrogan/stata/TwoPageStata.pdf
A discussion of pseudo R squared measures
https://stats.idre.ucla.edu/other/mult-pkg/faq/general/faq-what-are-pseudo-r-squareds/
Missing Data Methods
http://www.missingdata.org.uk/
A usedful paper on presenting modelling results
Connelly et. al (2016)
http://journals.sagepub.com/doi/pdf/10.1177/2059799116638002
Some useful Stata code for outputing model results
global path8 "e:\tables"
logit yvar i.x1 i.x2 x3
estimates store logit
#delimit ;
esttab logit using $path8\logit1.rtf,
cells("b(fmt(3))se(fmt(3)) t(fmt(3)) p(fmt(3))")
stats(r2_p N, fmt(%9.2f %9.0f) labels(R2(pseudo) n))
starlevels(* .10 ** .05 *** .01) stardetach
label mtitles(" ")
nonumber
title(Title of my model here)
nogaps replace ;
#delimit cr
More information from Ben Jann, Institute of Sociology, University of Bern, [email protected].
http://repec.org/bocode/e/estout/advanced.html
No research micro data will be distributed during this workshop.
The files required are available in this git repository.
Participants MAY NOT make copies of these files or distribute them.
The files have been specifically created by AQMEN for training and MUST not be used for social science research.
THE DATA ARE NOT REAL! DO NOT USE THEM FOR REAL ANALYSES!
We advise that participants galvanise their training experience and work with 'genuine' data files which they can dowload legally from the UK Data Service (http://ukdataservice.ac.uk/ - the old site is www.esds.ac.uk but you will be redirected).
At workshops where real survey datasets are used participants MUST sign a data access agreement.
Real administrative data are never used in training settings.