Statistics and graphics play

How can graphics and / or statistics be used to misrepresent data? Where have you seen
this done?

How can graphics and / or statistics be used to misrepresent data? Where have you seen
this done?

Statistics and graphics play a significant role in research as are used quantitatively to
define variability inherent in data, as well determine the probability of certain outcome, errors
and uncertainties. They are therefore required at all the stages of researches from the beginning
to the end. This paper delineates on how these statistics and graphics misrepresent data providing
a scenario that this has happened.
Some graphics and statistics are biased as they are based on individual opinions, views
while others are fabricated. One way that these graphics misrepresent data is the misconception
held by the users that graphics and statistics provide a measure of proof that certain information
is true. Statistics provide a measure of the probability after observing certain results and
therefore they require scrutiny to establish their margin of truth. Most statistics introduce
systematic errors in the data either intentionally or accidentally (Statistics Canada, 2014). The
person who analyses statistical information may not be knowledgeable or may not see these
errors hence leading to wrong conclusions. This happens especially when no supportive data or
information explains what the statistics or graphics mean. In such scenarios, such statistics may
be taken as fact instead of probability.
Another way that statistics and graphics can be used to misrepresent data is the situation
where by the source of the statistics and graphics is not factual or credible (Gardenier & Resnik,
2009). This leads to a misleading and biased statistics based on wrong information and when

such graphics and statistics are published, it leads to wrong information. The fact that statistics
deals with numbers, they are more convincing and many people may approach this information
with less skepticism without understanding or putting into consideration the fact that they may be
Graphics and statistics can be misrepresented by personal bias, inaccurate statistics and
also fictional data. For instance, a person that is opposed to a certain situiaon may interfere with
the statistics by changing the figures to suit his or her self-interest. It is very easy to manipulate
statistics without other users realizing this. This is especially experienced in situation that deals
with huge chunks of data.
Furthermore, these misrepresentations result from the sampling and non-sampling errors
in survey (Statistics Canada, 2014). Sampling errors happens because, a portion of the entire
population is studied not the entire population. This therefore leads to generalization of
statistics, which may not in reality represent the interests, or the opinions of the entire
population. Non-sampling errors are present in both censuses and sample surveys as they are not
restricted to the small size selected. Such errors may involve lack of knowledge on the part of
individuals analyzing the data and such lead to misrepresentation of the graphics and statistics.
Inappropriate use of these graphics and statistics in general can be traced in data analysis, data
collection, the design of the experiments, analysis methods and interpretation (Ercan et al. 2009).
Various examples exist where there has been data misrepresentation. A good example
was the case of Enron Company whereby managers falsified or misrepresented the figures in the
books of accounts misrepresenting the public on the financial viability of the company. When his
was revealed, the company had to collapse as it was declared bankruptcy. These instances are

also experienced during elections. Many of the election petitions arise from the premise that
there was data misrepresentation/falsification of statistics that favored the opponent candidates.
Similarly, in statistics there are various instances where by researchers conducted in a given
location come up with different results. These are enough examples that indicate that indeed
statistics and graphics may be misrepresented.




Ercan et al. (2009). Misuse of statistics in medical research. Eur J Genr Med, 4(3):128-134.
Gardenier, J., & Resnik, D. (2009). The Misuse of Statistics: Concepts, Tools, and a Research
Agenda. Accountability in Research: Policies & Quality Assurance, 9(2): 65
Statistics Canada. (2014). Misinterpretation of statistics.