CHAPTER 10- EVALUATING AND INTERPRETING INFORMATION
Evaluating you information begins with getting multiple views on the subject in question, doing this will help limit the amount of bias when you go to interpret what you've learned. Check the dependability of your sources: are they well supported? is it verifiable? is it believable? Take into account the weaknesses of each article's argument and if all their claims are sound.
Not all sources are equal check the timing- make sure you're up to date but if your research calls for it look in to historical findings as well, see how opinions and options have changed over the years. Published books, even new ones can contain information more then a year old. If it's a printed source, check the publishing information, check the author's credibility, as well as the publishers- was it published by a university or society etc. check the bibliography, citations can be a good source of other resources. If it is an online source, check it's credibility as well. Look in to the persons sponsoring the studies to see possible bias and motives. cross check sources.
In evaluating evidence, see if what you have is sufficient to prove your point, separate hard and soft evidence- factual, expert opinions from speculation. Is the evidence balanced and reasonable? is the data being interpreted for us? do they cloud their findings with opinions and spicy language?
Now we get to interpreting your findings, your overall judgement and conclusions about you research.
1) identify certainty level
ultimate truth- conclusive answers
probable answer- what is most likely true, new discoveries may challenge or change what we think
inconclusive answer- not enough is known of the subject, the answer is harder to see then thought
2) examine the underlying assumptions
some assumptions are the wide spread applications of sample groups
3) personal bias
4) other possible interpretations
Try to avoid error in your reasoning, " we derive conclusions about what we don't know y reasoning from what we do know [Hayakawa 37]" (P. 158).
1. faulty generalizations
2. faulty casual reasoning
3. faulty statistical reasoning
sanitized statistic- manipulated numbers
meaningless statistic- exact numbers used when approximations should be
undefined average- mean, median, mode confused
distorted percentage figure- no explanation of the original numbers
bogus ranking- items compared on ill defined criteria
correlation and causation mix up- correlation doesn't mean causation
biased meta-analysis- study of studies
fallible computer model- depend on assumptions
misleading terminology- expert and laypersons misinterpretation
Research has limits. validity and reliability- surveys can be misconstrued with people picking the answer they think is wanted. Flaws in research studies, there are limitations to all the types of studies. Measurement errors, all measurements are prone to error. deceptive reporting, the language used in the reporting will change the way people interpret your reports.
No comments:
Post a Comment