"Even
the most elementary statistical methods have their practical
effectiveness limited by measurement variation." (p. 46)

"If we are going to claim that our methods really help explain the world around us, it is important that we carefully discuss their limitations and implications." (p. 48)

In most research, one or more outcome
variables are measured.
Statistical analysis is done on the outcome measures, and conclusions
are drawn from the statistical analysis. Sometimes outcome measures are
answers to questions, so care needs to be taken in wording of
questions. The statistical analysis
itself often involves another choice of measure, often called a summary statistic. One common
source of misleading research results is giving inadequate attention to
the choice of either outcome variables or summary statistics.
Making a
good choice depends on the particulars of the context, including the
research question. Although there are some guidelines, there are no
one-size-fits-all rules. So aspects of this topic
can best be approached by examples."If we are going to claim that our methods really help explain the world around us, it is important that we carefully discuss their limitations and implications." (p. 48)

Stephan B. Vardeman et al, Elementary
Statistical Methods and Measurement Error, The American Statistician, vol. 64,
February 2010, pp. 46 - 51. ^{1}

1. This article offers suggestions for incorporating attention to measurement in statistics classes.

2. Most of the discussion applies to predictor variables as well as outcome variables.

Last updated May 12, 2011