Research Methods Flashcards
Why do we do a peer-reviews
- Keeps research honest
- stops poorly conducted research being accepted as fact
- makes people think about their methodology
- assess universities
- suggests improvemnts
- can help allocate funding
How are peer reviews done
- look at the methodology
- data analysis
- look at justification
Disadvantages of peer reviews
- hard to find experts
- sometimes it can be hard to find annimoity (if there are only a few people in the field) could lack objectivity
- there may be publication bias (may not want to publish negative results)
- sometimes may not want to publish research that contradicts the status quo
What are the two types of reliability?
Internal: in an interview do both researchers obtain the same results (inter-rater reliability)
external: how much results differ for example if you do the experiment will you get the same result each time
How can we test whether we have high reliability
Inter rater relibality
- compare the results of two researchers when they are looking at the same interview/observations. If they have the same results then the method has high inter-rater reliability.
- To improve you could look at how you opernalisatised, for example changing behavioural categories
test re-test: a test should give the same results twice. to improve, you can alter tests to improve the correlation
Split half method: half questions and ppts and see if people get same results, if not remove problematic questions
What is internal validity
has the IV changed the DV or as if it something changed the DV like extraneous variables or confounding variables
What is external validity
can you apply the findings to the public or day to day life (generalisability)
What can affect the internal validity of a study
- particapant variables (demand charctertics, personality, age)
- Lack of control (order effects, investigator effects)
- situational variables (temp, room size)
- Researcher Bias (lack objectivity)
What is face validity
does it claim to measure what its measuring,
What is ecological validity
when you can generalise to a different place or setting
What is mundane realism
is the task similar to what we would do in real life
What is population validity
The ability to generalise the findings to the wider population
What is temporal validity
Can you generalise to a different century, or decade
What is concurrent validity
the extent to which the test produces another established measure
e.g two tests of IQ should produce the same result
How would we improve our external validity
Using field study, natural observations etc
How would you do a sign test
- subtarct the two varibales
- then to work out s look for the least frequent sign
- to find n count all the +,- expect the 0s, this is N
- then using N you look for whether your results are significant
What is a type 1 error
This is when you accept your experimental hypothesis when you should have rejected it as the results were affected by random variables
How do we stop a type 1 error from occuring
we use a 0.01 level of signifcance
What is a type 2 error
You accepted your null hypothesis when you should have rejected it,due to you thinking there was no significance between variables
How do we prevent a type 2 error from happening
You use a 0.05 level of significance
what is nominal data
data we can set into differenet categories (gender, dogs,)
we can have a frequency count
What is ordinal data
It can be ordered but we are not sure the exact value between each point is the same
small -large (height)
What is interval data
when you can meaure data with equal sets of intervals (length, time, temp, )
What is the nnemonic for the stats table
S-pace (sign test) W-eather (wiliconxon) R-eally (related t test) C-contains (chi-squared) M-any (mann whitney) U-FOs (unrelated t test) C-hasing (chi squared) S-mall (spermans rho) P-igs (pearsons)