Spurious effects in random-intercept cross-lagged panel models: Results from simulations and reanalyses of data on self-esteem and problematic eating behaviors used by Beckers et al. (2023)
Published: Jan. 11, 2024
The
random-intercept
cross-lagged
panel
model
(RI-CLPM)
is
an
extension
of
the
traditional
model.
In
RI-CLPM,
prospective
effects
are
estimated
within
rather
than
between
individuals.
present
simulations,
we
found
that
RI-CLPM
susceptible
to
spurious
findings
when
observed
scores
on
two
variables
affected
by
common
auto-correlated
state
factors.
reanalyses
empirical
data,
contradictory
decreasing,
increasing,
and
null
within-individual
levels
self-esteem
problematic
eating
behaviors
among
Dutch
teenagers
(N
=
1856).
These
indicated
previously
reported
probably
were
spurious.
Caution
advised
using
as
it
may
produce
misleading
results.
Researchers
recommended
validate
alternative
analyses,
particularly
examining
person-mean
centered
scores.
Language: Английский
In aqua veritas: Triangulation of cross-lagged effects for improved causal inference
Published: May 14, 2024
Objective:
To
show
which
adjusted
cross-lagged
and
concurrent
effects
to
expect
from
panel
models
in
the
case
of
true
causal
increasing
decreasing
effects.Design:
Full
factorial
prospective
situ
experimental
study.Setting:
The
first
author’s
kitchen.Participants:
Twenty
plastic
cups.Main
outcome
measures:
Total
weight
(i.e.
including
their
content)
cups
before
after
adding
or
removing
water
to/from
them.
Results:
We
found
amount
added
have:
(1)
A
positive
effect
on
subsequent
total
when
adjusting
for
initial
weight;
(2)
negative
(3)
-
difference
cups.
Removing
filled
with
had
opposite
effects.Conclusions:
same
combination
signs
as
should
be
expected
between
studied
constructs
all
areas
research,
such
psychology.
Estimation
three
these
effects,
instead
only
one,
allows
stronger
inference
about
causality
is
recommended
order
triangulate
scrutinize
findings
models.
Language: Английский