monfils 418_Jan 23 2014

Humanratersoperatorsmaygetredbored soluon

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: ndent
variable
 could
affect
the
experimental
outcome
 •  Solu$ons
to
this
possible
problem:


 –  try
to
minimize
the
presence
of
these
other
factors
 –  if
they
are
unavoidable,
try
to
spread
their
effects
 equivalently
between
groups

 Categories
of
such
factors
have
been
iden$fied
 Threats
to
internal
validity
 •  Threats
to
general
experimental
designs
 –  Instrumenta$on
 –  Selec$on
 –  Mortality
 •  Threats
specific
to
pre‐
post‐test
designs
 –  History
 –  Matura$on
 –  Regression
to
the
mean
 Instrumenta$on
 •  The
reliability
of
an
instrument
(to
measure
the
 dependent
variable
or
manipulate
the
 independent
variable)
may
change
 •  Human
raters/operators
may
get
$red/bored
 Solu$on:
 
‐
monitor
equipment
 
‐
assess
interrater
reliability
 
‐
score
data
in
random
order
 Example:
abduc$on
experiment
 
Experiment
to
test
whether
watching
an
 educa$onal
video
can
teach
children
to
say:
 “No,
I
have
to
go
ask
my
teacher”
when
they
 are
asked
to
follow
a
stranger.
 Poche,
Yoder,
&amp...
View Full Document

Ask a homework question - tutors are online