week 8- critical analysis Flashcards
what are 7 features of a good piece of research
- Published in a good journal
- Large sample size
- Open science
- Pre-registration
- Peer reviewed
- Reliable methods
- Replicable
what does something being reliable mean
if we measure the same thing more than once we get the same answer
what does accuracy mean
whether it gives us the correct answer
what does validity mean
something measures what it intends to
why is replicability difficult
societies change over time
2 cons of observations are
- Some behaviours are difficult to observe as they are rare or private
- They can’t infer causal relationships
pro of observations
high ecological validity
what is open science
Making data and analysis script available
what is a survey
Lots of people asked
questions about their
thoughts or behaviour
pro of survey
can measure unobservable behaviour
2 cons of surveys
- people may give inaccurate responses, reducing ecological validity
- difficult to infer causal relationships
what are experiments
Generate conditions
that differ in only one
respect and investigate impact of
these manipulations
on another variable
pro of experiments
Can examine a causal relationship
(because we manipulated the IV)
cons of experiments
participant effects, experimenter effects, reduced external validity
define participant effects
Participant performs in the way they
think the experimenter want them to,
or in a way that is more socially
desirable
define investigator effects
Effect of
experimenter
on responses
how can we solve experimenter effects
using a double blind procedure
what is external validity
the extent to which conclusions from the study can be generalised
what is internal validity
The observed effects are only due to
the IV. Not due to chance or any other
factor.
how can we improve internal validity
random selection of ps, random allocation to conditions
can we always trust published research
No. Something published in a
peer-reviewed journal means
that other a couple of scientists
judged it suitable for publication, however they may be biased or not spot errors
what are the 3 tasks of critical appraisal
- Establish the facts:
– What did the researcher do? - Evaluate the article against its central
themes, with a balanced commentary:
– Does the data support the article’s conclusions? - How would you improve it?
how should we treat situational variables
Participants in all treatment groups should, as far as possible, have the same experience
how should we treat instrumentation variables
The same measuring instrument should be used for all groups
how do we treat nuisance variables
Random allocation of participants to
treatment groups
how do we treat order effects
Counterbalancing order
