Description

Best lists a series of questions to ask about a statistic reported in the media or literature. These questions can help the reader avoid several pitfalls in dealing with statistics. The author is from the University of Delaware.


 

A reader can be classified as:

(1) naïve: accepts a statistic without question

(2) cynical: rejects a statistic without question

(3) critical: reserves judgment until questions can be answered

 

Some of the questions that might be asked about a statistic:

(1) What are the possible sources for the statistic?

(2) How can that number be produced?

(3) Who produced the statistic?

(4) What are the interests of the person who produced the number?

(5) How are the key terms defined?

(6) What are alternative ways in which key terms might be defined?

(7) How can the phenomenon in question be measured?

(8) What sample was evaluated?

(9) How does the choice of the sample affect the results?

(10) Is the statistic being properly interpreted?

(11) If comparisons are being made are they appropriate?

(12) If an opponent reports a different statistic, how was it determined and why was that method chosen?

(13) Can apparent differences in statistics shown by 2 groups be due to how each group handled the data?

 


To read more or access our algorithms and calculators, please log in or register.