Analysis of Statistical Means
Research on a population can result in a large number of rating data, which may be difficult to understand. Therefore, a statistic is needed to represent the whole set of numbers.
The mean, or average is the most powerful way to represent a group of numbers, especially when about 25 or more people are surveyed. But what does the mean tell us? What standard, if any, do we have to make the results of the ratings more usable?
Research textbooks lack reference to methodology after the ratings are collected. The article Analysis of Statistical Means: A Proposed Standard [PDF] proposes a standard by which the results of numerical ratings, exemplified by the mean, can be easily understood and interpreted to answer the initial purpose of the research. The article analyses the results from a study that involved a large amount of data collected over a 13-year period, and proposes a standard for analysing the statistical means.
The Analysis of Means Standard provides a method for making all those rating numbers meaningful. When all questions being rated by a group of people are plotted against the Standard the message within the data is revealed. If this Standard is used over a period of time the variance from one rating to another becomes evident.
Analysis of Statistical Means: A Proposed Standard by Kerm Carlson is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.
Based on a work at www.researchvalues.com.
Permissions beyond the scope of this license may be available at http://www.researchvalues.com.