BEGIN:VCALENDAR
VERSION:2.0
PRODID:talks.ox.ac.uk
BEGIN:VEVENT
SUMMARY:Advanced Quantitative Methods\, Trinity Term 2023 - Professor Lars
-Erik Malmberg (University of Oxford)\, Dr Ariel Lindorff (University of O
xford)\, Dr Chris Wheadon (University of Oxford)
DTSTART;VALUE=DATE-TIME:20230503T093000
DTEND;VALUE=DATE-TIME:20230526T130000
UID:https://talks.ox.ac.uk/talks/id/59593bd0-205d-4829-95ab-6d30871c3781/
DESCRIPTION:In Trinity Term 2023 we will offer four in-person advanced qua
ntitative methods courses at the Department of Education\, University of O
xford. The first two course (multilevel and structural equation modelling)
are open for all interested Department of Education students and members
of staff. The later two courses (psychometrics and assessment analysis\, a
nd advanced modelling in R) have a limited number of places available for
Masters and Doctoral students. \n \nThe courses require a basic understand
ing of multiple regression modelling or other multivariate techniques. We
will use a variety of software\, including R-studio\, various R-modules\,
and Mplus (mostly the free demo-version) during the courses. \n\nStudents
and staff are welcome to sign up using the following link https://oxforded
ucation.eu.qualtrics.com/jfe/form/SV_8faQDVnVPLeJSvk. \n\nOur program is:
\nOxford TT Week 2 Multilevel modelling (Ariel Lindorff) \nOxford TT Week
3 Structural equation modelling (Lars Malmberg):\nOxford TT Week 4 Psychom
etrics and assessment analysis (Chris Wheadon)\nOxford TT Week 5 Advanced
models in R (Lars Malmberg)\n\nOxford TT Week 2\nMultilevel modelling (Ari
el Lindorff): Open to all students (and staff) in the Department of Educat
ion (Wed 3 May 2023\, 9:30-12:30). We will meet in Seminar Room D (with on
line access as an alternative to joining in person). This workshop will bu
ild on participants’ existing knowledge of regression models\, introduci
ng multilevel models for hierarchically nested data. We will be working in
R\, primarily relying on the lme4 package. We will cover fixed\, random a
nd contextual effects\, and learn how to visualise findings. \n\nOxford T
T Week 3\nStructural equation modelling (Lars Malmberg): Open to all stude
nts (and staff) in the department. We will convene 9-12 in person in Semin
ar Room D (with online access). Pre-recorded materials and example dataset
s and models (in R and Mplus) will be made available in advance. See furth
er details of each day below. \n\nTue 9 May\, 9-12 Intro Structural Equat
ion Models (SEM)\nWed 10 May\, 9-12 Longitudinal SEM\nFri 20 May\, 9-12
Multilevel and intraindividual SEM\n\nOxford TT Week 4\nPsychometrics and
assessment analysis (Chris Wheadon): Up to seven departmental Masters and
Doctoral students can audit the course. Participants will be selected base
d on closeness between own research and the course contents. Three major f
rameworks will be discussed and compared – Classical Test Theory\, Compa
rative Judgement and Item Response Theory. The difference between the Brad
ley-Terry-Luce model as well as the 1-\, 2- and 3-parameter logistic (PL)
models will be examined. For IRT\, the 1-PL model widely known as the Rasc
h model will be the focus. Advantages and disadvantages of the frameworks
and different models will be overviewed\, and implications of model choice
will be discussed. Teaching will be in seminar rooms K/L from 9.30 to 5
on Monday (15/5)\, Tuesday (16/5) and Thursday (18/5)\, and 9-30 to 12-30
on Wednesday (17/5) and Friday (19/5). \n\nOxford TT Week 5\nAdvanced mode
ls in R (Lars Malmberg 22-26/5/2023): Up to seven departmental Masters and
Doctoral students can audit the lecture part of the course (each morning
22-26/5). Participants will be selected based on closeness between own res
earch and the course contents (see further details of each day below).\n\n
Students and staff who wish to engage in self-learning can have access to
Canvas in which both the Intro QM and Intermediate QM courses now have ext
ensive R (and SPSS) lecture and application materials (videos). Students c
an get access to all LM's and AL's advanced course materials for self-lear
ning too.\n\n \n*Week 3 of Trinity Term Structural Equation Models*\n \nTu
e 9/5\, 9-12 in Seminar Room D\, possibility to join online\nIntroduction
to Structural Equation Modelling \n \nPre-recorded videos and materials ma
de available in advance of the session \n \nThe concept of a latent constr
uct is central in the social sciences. A latent construct is a non-directl
y observed phenomenon (e.g.\, attitude\, socioeconomic status) that we can
model using manifest (observed) variables (e.g.\, survey and questionnair
e responses\, observation scores)\, by partitioning out residual (i.e.\, u
niqueness\, error variance). The structural equation model (SEM) is divide
d into two parts. In the measurement part of the model\, we can inspect wh
ether manifest variables measure the constructs they are intended to measu
re. This model is called confirmatory factor analysis (CFA) which allows t
he researcher to test whether an a priori model fits data\, and whether th
is also holds across multiple groups. If measurement is satisfactory\, the
relationships between constructs can be estimated in the structural part
of the SEM. Complex relationships between manifest variables and/or latent
constructs can be tested in path-models not possible to specify in the mu
ltiple regression framework. During the course we will cover worked exampl
es relevant for educational\, psychological and social sciences. \n \nPre
-requirements \nParticipants need to understand the basics of multiple reg
ression\, or other relevant multivariate statistics. \n \nContents \nVideo
-clip 1 Basic concepts\, models and measurement. From multiple regression
to path-models using manifest variables. \nVideo-clip 2 Observed (manifest
) variables and unobserved (latent) constructs. Specification of measureme
nt models for testing quality of measurement\, using continuous and dichot
omous manifest variables. Goodness-of-fit indices. \nDemonstration video-c
lips. Models in R-lavaan \nDemonstration video-clips. Models in Mplus \n \
nProgram \n09.00-09.30 Recap of basics of SEM (questions can be posted in
the Q&A document) \n09.30-10.30 Measurement models and goodness of fit \n
10.30-10.45 Break \n10.45-12.00 Structural models for answering substantiv
e research questions \n \nSoftware: We will mainly use the Mplus demo htt
ps://www.statmodel.com/demo.shtml) software. Parallel code is available in
R (Lavaan) (http://lavaan.org). Materials are made available in advance o
f the session\, and can also be used for self-study. \n \n \n \nWed 10/5\,
9-12 in Seminar Room D\, possibility to join online\nStructural Equation
Modelling of longitudinal data \nPre-recorded videos and materials made a
vailable in advance of the session \n \nIn this follow-up of the introduct
ion to SEM course we focus on SEM for longitudinal data. Prospective longi
tudinal data is typically collected over longer periods of time e.g.\, ter
ms or years. Using SEM we can model repeated latent constructs over time u
sing autoregressive models\, that is a construct at the concurrent time-po
int regressed on that construct at a previous time-point. We can also test
whether the measurement is invariant (measured in the same way) across th
e time-points. When particular interest is in individual differences in ch
ange over time\, we can model time explicitly in the latent growth curve m
odel. The worked examples are based on educational longitudinal data\, rel
evant for social sciences. \n \nPre-requirements: \nParticipants need to
understand the basics of multiple regression\, other relevant multivariate
statistics\, and have some exposure to either regression or SEM. \n \nCo
ntents \nVideo-clip 1: Introduction to longitudinal (repeated measures mod
elling) \nVideo-clip 2: Auto-regressive modelling \nVideo-clip 3: Growth
modelling \n \nWednesday 18 May \n09.00-09.30 Recap of longitudinal model
ling (questions can be posted in the Q&A document) \n09:30-10:30 Autoregr
essive modelling\, and testing of measurement invariance \n10.30-10.45 Bre
ak \n10.45-12:00 Growth models \n \nSoftware: We will use R Lavaan and Mp
lus software (full version\, input and output files are available). Materi
als are made available in advance of the session\, and can also be used fo
r self-study. \n \n \n \nFri 12/5\, 9-12 in Seminar Room D\, possibility t
o join online\nStructural Equation Modelling for multilevel and intraindiv
idual data \nPre-recorded videos and materials made available in advance o
f the session \n \nMultilevel structural equation modelling (MSEM) combine
s the best of two worlds\, the multilevel model (MLM) and SEM. The multile
vel SEM (MSEM) allows us to test structural validity in two or more hierar
chical levels\, and specify level-specific associations between level-spec
ific predictors and outcomes. MSEM can be applied to different hierarchica
l data structures quite commonly found in educational research\, e.g.\, st
udents nested in classrooms\, or time-points nested in students. \nIn the
first session we introduce multilevel modelling using manifest indicators
\, comparison of notation in MLM and MSEM\, and model specification in the
SEM framework. \nIn the second session we specify latent constructs for
intraindividual data (time-points nested in students) and include level-sp
ecific predictors. We specify fixed and random effects models assuming “
individuals as their own controls” type of models\, in which the time pe
rspective is not specified (Malmberg\, 2020). \nIn the third session we a
pply Dynamic SEM assuming stationarity (no mean trends over time)\, specif
ying equidistant time-lags for lagged variables in diary data (a working-l
ife dairy for a year). These time-series like models can be specified usin
g the Bayesian estimators\, allowing us to investigate within-person varia
bility. \n \nContents \nVideo-clip 1: Introduction to multilevel modelli
ng in SEM\, multilevel factor structures \nVideo-clip 2: Intraindividual
SEM \nVideo-clip 3: Dynamic SEM \n \nFriday 12 May \n09.00-09.30 Recap o
f multilevel SEM (questions can be posted in the Q&A document) \n09:30-10:
30 Work on MSEM and ISEM \n10.30-10.45 Break \n10.45-12.00 Work on DSEM
\n \nSoftware: We will mainly use the Mplus software (full version\, input
and output files are available\, demo (https://www.statmodel.com/demo.sht
ml). Some parallel code for plotting is available in R (Lavaan). Materials
are made available in advance of the session\, and can also be used for s
elf-study. \n \n \n \n*Week 5 Trinity Term*\n \nMon 22/5 to Fri 26/5\, 9-1
2 in Seminar Room D \nAdvanced models in R (Lars Malmberg 9-12\, 22-26/5/2
023): Up to seven departmental Masters and Doctoral students can audit the
lecture part of the course (each morning 22-26/5). Participants will be s
elected based on closeness between own research and the course contents. \
n \nIn this course we will introduce relevant packages in R for regression
models. We will use both existing secondary data and simulated data for l
earning key concepts for specifying models and interpreting findings. \n
\nMon 22/5\, 9-12 (The regression model) \nWe start with an overview of mu
ltivariate statistics. Introduction to descriptive statistics and regressi
on modelling for continuous dependent variables\, using R. We will cover b
asic concepts: model\, notation(s)\, estimation techniques and visual insp
ections. We will inspect continuous and dummy-coded predictors\, inspect r
esiduals and other indices of model-health. \n \nTue 23/5\, 9-12 (Expan
sions of the regression model) \nWe will expand the regression models to e
xamples of interaction-effects\, moderation models\, and mediation models\
, also using path-analyses in R Lavaan. Link-functions for logistic regres
sion (for binary outcomes) and probit regression (for ordinal data) are pr
esented. \n \nWed 24/5\, 9-12 (Introduction to path-analysis and structur
al equation modelling) \nIntroduction to path-models and measurement model
s in R Lavaan. We will specify measurement models for continuous indicator
s of latent constructs\, as well as measurement models for dichotomous ind
icators (0 = incorrect\, 1 = correct) as indicators of latent constructs\,
analogous to IRT models. \n \nThurs 25/5\, 9-12 (Multilevel regression)
\nWe will expand the regression models to multilevel regression analysis f
or hierarchically nested data\, starting with students nested in classroom
s in R lmer and Lavaan. We will cover fixed\, random and contextual effect
s\, and learn how to visualise findings. \n \nFri 26/5\, 9-12 (Advanced m
ultilevel regression) \nWe will expand the multilevel regression model to
include other hierarchical data-structures: time-points nested in persons\
, test-items nested in assessors\, specifying these as growth models\, mul
tivariate models and multiple membership models. \n\nSpeakers:\nProfessor
Lars-Erik Malmberg (University of Oxford)\, Dr Ariel Lindorff (University
of Oxford)\, Dr Chris Wheadon (University of Oxford)
LOCATION:15 Norham Gardens\, 15 Norham Gardens OX2 6PY
TZID:Europe/London
URL:https://talks.ox.ac.uk/talks/id/59593bd0-205d-4829-95ab-6d30871c3781/
BEGIN:VALARM
ACTION:display
DESCRIPTION:Talk:Advanced Quantitative Methods\, Trinity Term 2023 - Profe
ssor Lars-Erik Malmberg (University of Oxford)\, Dr Ariel Lindorff (Univer
sity of Oxford)\, Dr Chris Wheadon (University of Oxford)
TRIGGER:-PT1H
END:VALARM
END:VEVENT
END:VCALENDAR