Manipulation of statistics
in information: a worrying practice
Manipulación de la
estadística en la información: una práctica preocupante
José Luis Hidalgo Torres*
ABSTRACT
Statistics
is a very versatile and indispensable tool for analyzing and making decisions
in a wide range of areas, such as science, politics and communication. However,
it can be manipulated for questionable purposes and generate serious
consequences in society as a result of its use. The manipulation of statistics
in information produces a distortion of the truth, affecting people's
decision-making, individually and in groups, producing: making erroneous
decisions, based on false information; economic losses due to investments or
purchases based on unfounded information; and, health risks, caused by the lack
of important information on safety or effectiveness. To illustrate the case, a
methodology based on critical analysis was applied; evaluating and contrasting
sources, identifying biases and inconsistencies and selecting validated
information, conducting a context search, examining the environment and the
variables that influence the statistics and the techniques used to display the
data at the convenience of the interested parties. Finally, the ethical and
social implications of this manipulation were discussed, as well as possible
solutions to prevent and combat it.
Keywords:
Manipulation, distortion, consequences, critical analysis, public perception,
ethics.
RESUMEN
La estadística es una herramienta
muy versátil e indispensable para analizar y tomar decisiones en una amplia
gama de ámbitos, como la ciencia, la política y la comunicación. Sin embargo, puede
ser objeto de manipulación con fines cuestionables y generar graves
consecuencias en la sociedad, como producto de su uso. La manipulación de la
estadística en la información produce una distorsión de la verdad, afectando la
toma de decisiones de las personas, de manera individual y grupal, produciendo:
toma de decisiones erróneas, basadas en información falsa; pérdidas económicas
por inversiones o compras en base de información infundada; y, riesgos para la
salud, provocados por la falta de información importante sobre seguridad o
eficacia. Para ilustrar el caso, se aplicó una metodología en base al análisis
crítico; evaluando y contrastando las fuentes, identificando sesgos e
inconsistencias y seleccionando información validada, realizando una búsqueda
de contexto, examinando el entorno y las variables que influyen en las
estadísticas y las técnicas utilizadas para mostrar los datos a conveniencia de
los interesados. Finalmente, se discutieron las implicaciones éticas y sociales
de esta manipulación, así como posibles soluciones para prevenirla y
combatirla.
Palabras clave: Manipulación, distorsión,
consecuencias, análisis crítico, percepción
INTRODUCTION
Statistics is a tool that allows organizing, analyzing
and interpreting data, becoming a fundamental discipline to be used in a wide
range of fields, which could start with science, administration, government
management, politics, media, etc.
The article analyzes the impact of statistical
manipulation on public perception and explores the techniques used to distort
statistical data. Manipulation of statistical information is a serious problem
that threatens trust in information, informed decision making, and social
welfare. It is crucial to investigate this issue to better understand its
mechanisms, consequences and possible solutions.
· This
practice can be carried out in a variety of ways, such as, for example:
- Selecting the data to be presented. One can choose to present only data
that support the position one wishes to defend, leaving aside data that do not
(Kuhn, Anahid, & Modrek, 2021).
- Modify the data. One can alter the data, so that they appear more
favorable to the position one wishes to defend.
.
- Present the data in a misleading way. Graphs, tables or indicators may
be used that distort the meaning of the data.
Consequently, what we want to emphasize is the
importance of transparency, honesty, individual and collective responsibility
in the presentation of statistical data, to ensure that a well-informed society
is able to make decisions based on evidence.
MATERIALS
AND METHODS
For
the realization of this article we used as specific inputs, information
provided by relevant scientific articles that identify how statistical
manipulation is produced and applied; and statistical information on the
sectorization of the employed population (15 years and older) between Sep. 20 -
Nov. 23 in Ecuador, obtained from the INEN web pages.
Qualitative method: The application of this method was used to obtain the
criteria, opinions of the scientific articles and others, from the web pages
that support the structural sequence of the content of this article that refers
to statistical manipulation.
Quantitative - descriptive method: The use of this method was used to show an
applied example of how statistical data can be shown and interpreted, according
to the applied procedure.
First, the motivations behind the manipulation of
statistics, which may include political, commercial or ideological interests,
etc., are examined.
Then, concrete examples of how this manipulation is
carried out in different fields, such as economic reports, scientific studies
and opinion polls, are given. Techniques
such as cherry picking, misleading presentation of graphs and omission of
relevant information are explored.
In addition, the consequences of the manipulation of
statistics on public perception are analyzed. This practice can generate confusion and
misinformation, making it difficult for citizens to make informed decisions.
The ethical and social implications of this manipulation are also discussed,
including the damage to public trust and the erosion of the credibility of
information sources.
And an application is described, to show how
statistical information can be displayed with the use of various procedures, to
obtain certain results that are not visualized in the way they are reported to
the public.
The manipulation of
statistics is an increasingly common practice in the media. The media can use
the manipulation of statistics to attract the attention of readers or viewers,
or to promote a certain agenda.
Some examples of manipulation of statistics in the
media are:
- Presenting data without context. Data
may appear more relevant or impactful if presented without context . For
example, it may be said that the number of crimes has increased by 20%, but if
the time period in which this increase occurred is not specified, the data can
be misleading.
- Using misleading graphs or tables. Graphs or tables can be used to
distort the meaning of the data. For example, a graph with an
exaggerated scale can be used to make a change appear more important than it
really is.
- Using misleading language. The language used to
present data can be misleading. For example, the word reduction may be used to
refer to a 1% decrease, when in fact it is a very small change.
The manipulation of statistics can have negative
ethical and social implications, because it can be used to mislead people and
manipulate them into taking a certain position or making a certain decision.
Ethical implications
Manipulation of statistics can violate the following
ethical principles:
- Truthfulness. Information must be truthful and accurate. Manipulation of
statistics can distort reality and create a false picture of the situation.
- Transparency. People have the right to know the
information in a transparent and complete way. Manipulation of statistics can
hide information or present it in a misleading way (Banco Central Del Ecuador,
n.d.).
- Fairness. Information should be presented in a
fair and equitable manner (Economic Commission for Latin America and the
Caribbean, n.d.). Manipulation of statistics can be used to discriminate
against or disadvantage certain groups of people.
RESULTS
The manipulation of statistics can have the following
negative social consequences:
- Influence on public decisions. The manipulation of statistics can be
used to influence public decisions, such as government policies or laws (Aragão & Linsi, 2020). Being
able to have a negative impact on society, by making decisions based on
misleading information.
- Disinformation. The manipulation of statistics can be used to
disseminate disinformation, which is false or misleading information, with the
purpose of misleading people. Disinformation can have a negative impact on
society, generating distrust in institutions and making it difficult to make
informed decisions.
- Manipulation of public opinion. Manipulation of statistics can be used
to manipulate public opinion on controversial issues. This has a negative
impact on society, dividing it and making dialogue and consensus difficult.
Consequences of statistical manipulation
The manipulation of statistics can have serious consequences for society. For
example, it can be used to:
- Influence public decisions. Politicians can use
the manipulation of statistics to justify their policies or to attack their opponents .
- Promote products or services. Companies can use
the manipulation of statistics to sell their products or services.
- To induce public opinion. The media can use
manipulation of statistics to influence public opinion on controversial issues
(Anna, Kalinina, & Yusupova, 2019).
How to identify the manipulation of statistics.
It is important for citizens to be aware of statistical manipulation and be
trained to identify it. To do so, the following aspects should be considered:
- Be critical of the information you receive. Do
not get carried away by flashy headlines or data that seem to confirm certain
beliefs.
- Ask about the context: In what time period was
the data collected? What factors may have influenced the data?
- Analyze graphs and tables. What information are
they showing? How are they presented?
- Be suspicious of misleading language. What does
the language actually mean?
As a second aspect, statistical information provided on the INEC web page is
used to show an example of how statistical information can be manipulated by
applying different techniques that are valid from any point of view, but that
show different ways of looking at quantitative information and that can lend
themselves to different types of interpretation.
To begin to develop what has been indicated, some aspects of the procedure
followed are mentioned:
1. The statistical information below corresponds to a
file 202311_Tabulados_Mercado_Laboral_EXCEL.xlsx found in the section
Tabulations and historical series (Excel button) of INEN's web page; within
this file, the information chosen for manipulation is in the Excel sheet named:
4.
The information of 4. Sectorization of Employment,
consists of percentages of the Sectorization of the Employed Population (15
years and older) at the national level, urban and rural sectors, month by month
from June 2007 to November 2023 (Figure 1).
2. As a first form of manipulation of the statistical information, it was
chosen solely for the purpose of showing how the information can be manipulated
(Figure 2), to alter the perception of the population, with the information of
the formal and informal employment sectors, from September 2000 to November
2023, with two-month intervals.
Figure 1 Percentages
of the sectorization of the employed population (15 years and older) in
Ecuador.
Figure 2.
Labor market in Ecuador between Sep-20 to Nov-23.
Source: Information taken from
Interpretation:
Formal sector: Considering as a base 40% for all periods, it is visualized that
in the corresponding months of 2020 until the year 2021, during the COVID
pandemic, the labor market in Ecuador has fluctuated by about 42%; from the
month of July 2021 until November 2023, it has had a slight increase of
approximately between 1 - 2%, with small decreases in the months of January,
May 2022 and January 2023; considering all these elements, it could be stated
that the formal labor market in Ecuador, has almost been a constant.
Informal sector: In a very similar manner to that
indicated for the formal sector, we consider 50% as the basis for analysis,
with slight increases between 1 - 3% in the same months indicated in the formal
sector. This labor market could also be considered almost constant, but it
should be taken into account that it is not the ideal way to prioritize the
population's livelihood.
By showing the information in this way, using the two
market type variables, the scale on which the results are shown allows me to
disguise or smooth the variation suffered by the labor market in Ecuador.
Obviously, for this type of effect to occur, it depends on the structuring of
the state processes and its policies, financial management, the controls that
should be applied, etc.
3. As a second way of manipulating the statistical
information, two other more explicit ways of showing the statistical results
are shown through graphs (Figure 3), which give an understanding of aspects
that go beyond what is seen in (Figure 2), in relation to the formal
sector of the labor market in Ecuador.
Figure 3.
Graphs of the fluctuation bands of the formal labor market in Ecuador between
Sep -20 to Nov -23.
|
Límite de
control superior: |
LCS = Promedio +
3*Desviación |
||
|
Límite central: |
|
LCS = Promedio |
|
|
Límite de
control inferior: |
LCI = Promedio
- 3*Desviación |
||
In (Figure 3), two graphs are shown with two totally
different statistical procedures, but showing the real growth or decrease of
the labor market in Ecuador, but on an average basis for both cases. While in
(Figure 1), the graph shows a constant appearance, in these graphs it can
be seen that in only 7 or 8 months of the 20 months taken for analysis, these
are the months in which there has really been a low growth between 1 - 2% of
the labor market.
Therefore, the way in which the results are presented
does influence the appreciation that can be given to the information and
consequently, depending on personal, group and institutional interests, the
information can be manipulated according to the purposes pursued by each one.
4. An additional example could be the misinformation
of statistical data that occurred during the COVID pandemic in Ecuador, a
situation that occurred at the international level, when the media reported the
number of deaths, while the official media denied and reported much lower
numbers than what was actually reflected in the news.
One can also mention the rise of the internet, which has led to an increase in
disinformation, as fake news can spread quickly and easily. This has led to a
loss of trust in traditional media and has led to some people choosing to avoid
news consumption altogether.
Factors that have contributed to the rise of
misinformation include the emergence of citizen journalists, the growing
influence of commercial and ideological interests, and the loss of credibility
of traditional media.
Fake news has become an industry of its own, with
people paid to write sensational stories and misleading headlines. Social
networks contribute to its rapid spread.
The people most vulnerable to fake news are those with
the least access to verified information, such as people in developing
countries and teenagers and young adults.
DISCUSSION
The manipulation of statistics in information is a
worrying practice that undermines confidence in data and institutions. It is
essential to be aware of this problem and to promote transparency and honesty
in the presentation of statistical data. Only in this way will we be able to
guarantee an informed society capable of making evidence-based decisions.
Finally, possible solutions are proposed to prevent and combat the manipulation
of statistical information. One of the proposed solutions is to promote
statistical literacy among the population, so that people can understand and
critically evaluate the information that reaches them. It also highlights the
need to promote transparency in the presentation of data, so that they are
shown in a clear and accessible manner, without distortions or manipulations.
In addition, it is proposed to strengthen the mechanisms for verification and
evaluation of information, in order to be able to detect and refute statistical
manipulations.
Finally, the importance of individual and collective
responsibility in the consumption of information is emphasized. Each individual
must be aware of the possibility of manipulation and must take responsibility
for verifying and contrasting the information he or she receives. Similarly,
society as a whole must demand transparency and honesty in the presentation of
statistical data, in order to be able to trust the information and make
informed decisions.
REFERENCES
Anna, V., Kalinina, E. E., &
Yusupova, E. V. (2019). Means of Influence on Public Opinion in Political
Context: Speech Manipulation in the Media. Media Watch. doi:10.15655/MW/2019/V10I2/49625
Anónimo. (s.f.). Las fake news y las estrategias de
verificación del discurso público: Caso Ecuador Chequea. Obtenido de Las
fake news y las estrategias de verificación del discurso público:
https://revistas.usfq.edu.ec/index.php/perdebate/article/view/1558/1706
Aragão, R., & Linsi, L. (24 de 05 de 2020). Many shades of wrong: what
governments do when they manipulate statistics. Review of International Political Economy, 1 -26.
Banco Central Del Ecuador. (s.f.). Código de conducta:
Normas para la producción y difusión de estadísticas del Banco Central del
Ecuador. Obtenido de Banco Central Del Ecuador:
https://www.bce.fin.ec/codigo-de-conducta
Chi-Hong,
L. W. (2023). Managing statistical misinformation in marketing: A study on
consumers’ attitudes and capabilities. Asian Journal of Empirical Research. doi:doi: 10.55493/5004.v13i1.4748
Comisión Económica para América Latina y el Caribe.
(s.f.). Equidad, desarrollo y ciudadanía. Obtenido de CEPAL 2000:
chrome-extension://mhnlakgilnojmhinhkckjpncpbhabphi/pages/pdf/web/viewer.html?file=https%3A%2F%2Frepositorio.cepal.org%2Fserver%2Fapi%2Fcore%2Fbitstreams%2Fa5fcd682-bdec-4b63-9621-693d36c497f8%2Fcontent
INEC. (11 de 2023). Estadísticas Laborales – noviembre
2023. Obtenido de INEC:
https://www.ecuadorencifras.gob.ec/estadisticas-laborales-enemdu/
Jannetje,
E., Driessen, D. A., Vos, I. S., & Casper, J. A. (2022). Misleading
graphs in context: Less misleading than expected. PLOS ONE.
doi:10.1371/journal.pone.0265823
Kaviani, V.
J., Amiri, P., Zamani, F. B., & Akhlaghi, N. (2021). Modification Data
Attack inside Computer Systems. A critical Review.
Kuchyk, H.
(2019). Information manipulation in the media as an element of political
communication. doi:10.25264/2519-2558-2022-15(83)-3-6
Kuhn, D.,
Anahid, S., & Modrek. (2021). Choose Your Evidence. Science Education.
Orús, A. (04 de 01 de 2024). Noticias falsas y
desinformación en el mundo - Datos estadísticos. Obtenido de Statista:
https://es.statista.com/temas/10126/fakes-news-y-desinformacion/#topicOverview
Polo Santillán, M. Á. (2019). La responsabilidad ética. Scielo.
Obtenido de
https://www.scielo.cl/scielo.php?script=sci_arttext&pid=S0718-92732019000100049
Portaltic/EP. (15 de 01 de 2024). Los modelos de lenguaje
pueden entrenarse para mostrar un comportamiento engañoso, que además
persiste. Obtenido de Portaltic:
https://www.europapress.es/portaltic/sector/noticia-modelos-lenguaje-pueden-entrenarse-mostrar-comportamiento-enganoso-ademas-persiste-20240115130958.html
Pullinger,
J. (2020). Misuse of statistics: Time to speak out. Statistical journal of
the IAOS, 79-84.
Robledo, M. P. (02 de 11 de 2023). La desinformación y la
manipulación de la opinión pública: un análisis en profundidad. Obtenido de
El Analista:
https://elanalista.com.ar/la-desinformacion-y-la-manipulacion-de-la-opinion-publica-un-analisis-en-profundidad%EF%BF%BC/
Rodriguez-Alveal, F. E. (2017). Alfabetizacion estadistica
en profesores de distintos niveles formativos. Educação & Realidade, 19.
Obtenido de
chrome-extension://mhnlakgilnojmhinhkckjpncpbhabphi/pages/pdf/web/viewer.html?file=https%3A%2F%2Fwww.scielo.br%2Fj%2Fedreal%2Fa%2FXNPpMpCD6y78gbcxtyZRBpD%2F%3Fformat%3Dpdf%26lang%3Des%23%3A~%3Atext%3DPara%2520Garfield%252C%2520delMas%2520y%2520Chance%2Cco
Rodríguez-Cruz, I. (2013). Información ambiental y
respuesta social en España: análisis de los casos de Doñana y las políticas
energéticas en la prensa nacional y local. Obtenido de Semantic Scholar:
https://www.semanticscholar.org/paper/Informaci%C3%B3n-ambiental-y-respuesta-social-en-Espa%C3%B1a%3A-Rodr%C3%ADguez-Cruz/450bb3191bb1f9ea11247a9ba2fbd5a697cc40ca
Sanhueza, J. (14 de 02 de 2019). Los datos sin contexto.
Obtenido de Noticias UAI:
https://noticias.uai.cl/columna/los-datos-sin-contexto/
Sojo, C. (2022). El estado bajo escrutinio: opinión
pública, estatalidad y desempeño gubernamental en América Latina. Santiago:
Naciones Unidas.
Teixidó Cabezas, L. (21 de 02 de 2023). La importancia de
la transparencia en las organizaciones, la visión de SEIT S.A. Obtenido de
Pacto Mundial Red Española:
https://www.pactomundial.org/tribuna/la-importancia-de-la-transparencia-en-las-organizaciones-la-vision-de-seit-s-a/#:~:text=Ser%20transparente%20es%20dar%20informaci%C3%B3n,a%20nuestro%20equipo%20de%20trabajo.
Uzobo, E.
&.-O. (2019). Damned lies, half truth and social statistics. Damned
lies, half truth and social statistics. African Journal for the
Psychological Study of Social Issues, 22(1), 176-186.
Wallace, J.
(2013). Juking the Stats? Authoritarian Information Problems in China. Social Science Research Network.
*Mgtr. – Docente Instituto
Tecnológico Superior Corporativo Edwards Deming, jlhidalgot@hotmail.com https://orcid.org/0000-0002-4671-3116