The question here it would seem should focus more on the manipulation of data rather than that of adding participants to create change in results... (Or at least that's what I think)
The reason I have said this is due to the belief that in research another psychologist could easily find a larger sample and replicate a study, which may create a change in results BUT appears to still in keep with 'good' science. 'Good' science for the sake of this blog being honest untampered with research, or research that hasn't been tampered with to create a desired effect.
Manipulating methodology is all well and good when it comes to Psychology because that is what Psychology is about. Establishing causal relationships. If we don't test, we don't know. However to manipulate data is in my opinion, 'bad' science.
(Here is normally where a reference would go... There are no good ones)
To continue, manipulation of data seems to give off a sense of laziness on the researchers part. Understandably there may be problems, regarding factors such as funding to carry on research, yet it doesn't warrant manipulating data to create an effect that wasn't visible in the first place.
Does anyone agree/disagree/have a decent piece of research to back up or refute what I've said?
Comments appreciated.
Obviously having a larger sample will produce more reliable results. Replication is also good, for example if you had a study which used only 20 participants and got X result, although you shouldnt manipulate results, there is no reason the experement should not be replicated. If the study was done again and found to get similer results then it would be less likely the results were due to chance and that the independent variable did in fact cause the changes to the dependent variable.
ReplyDeleteThe simple adding of extra participants into a study in itself is, in my opinion, not 'bad' science; as leighanne pointed out, larger samples are more reliable.
ReplyDeleteThus, to add participants and find different results (a difference where there wasn't one) could simply be that, now that you have more participants, the data is reflecting the actual situation more accurately.
However, if one actually has the aim of getting a certain result, as is implied by adding participants until an effect is found, it calls into question whether the experimenter is actually behaving in an objective, unbiased way. Perhaps, if they are so keen to observe an effect, they are behaving in a certain fashion with their participants which could trigger an effect in and of itself. This is where, in my opinion, the 'bad' science begins.
I do agree that replicating using a larger sample size ( or the same size in a different setting/location) would be 'good' science because it does not manipulate effects unreasonably and this can be discussed within the research itself. However, manipulating already existing data to fit a hypothesis is 'bad' science which I do not agree with. By manipulating these results a false impression can be given and a theory which may be wrongfully accepted (or rejected) when other theories should be considered/ rejected (depending on the theory in question).
ReplyDelete