Mona Chalabi: 3 ways to spot a bad statistic
Mona Chalabi: Tres vías para detectar una mala estadística
Mona Chalabi tries to take the numb out of numbers. She's left with lots of "ers." Full bio
Double-click the English transcript below to play the video.
about statistics today.
a little bit wary, that's OK,
de inmediato, está bien,
kind of crazy conspiracy theorist,
de teorías de conspiración,
especially now, you should be skeptical.
actualmente, deberían ser escépticos.
distinguir números fiables
which numbers are reliable
some tools to be able to do that.
para poder hacer eso.
I'm talking about here.
les hablaré.
this anti-aging cream."
esta crema antiedad".
roll our eyes at numbers like that.
con números como esos.
are questioning statistics like,
estadísticas como,
en EE. UU. es del 5 %".
rate is five percent."
no proviene de una compañía privada,
it doesn't come from a private company,
distrust the economic data
desconfían de los datos económicos
del presidente Trump,
it's even higher;
in our society right now,
en nuestra sociedad actualmente,
with these government numbers.
la gente y los números del gobierno.
these statistics are crucial,
que estas estadísticas son cruciales,
of society as a whole
la sociedad como un todo
emotional anecdotes
son elitistas,
and they don't really reflect
y que no reflejan en realidad
in people's everyday lives.
de la gente.
is winning the argument right now.
va ganando la pelea hasta ahora.
of alternative facts,
this kind of common ground,
con estas estadísticas,
para el debate.
moves in the US right now
statistics altogether.
estadísticas gubernamentales.
about measuring racial inequality.
para medir la inequidad racial.
del gobierno no debe usarse
money should not be used
sobre la segregación racial.
políticas justas
current levels of unfairness?
los actuales niveles de injusticia?
sin datos sobre salud o pobreza?
on health or poverty?
sobre inmigración
about immigration
and leaving the country?
that's where they got their name.
de allí viene su nombre.
measure the population
los números del gobierno,
beyond either blindly accepting
que aceptarlos ciegamente
para detectar malas estadísticas.
to be able to spot bad statistics.
in a statistical department
de estadísticas
iraquíes se vieron forzados
had been forced from their homes
como resultado de la guerra,
but it was also incredibly difficult.
pero muy difícil también.
of our numbers --
de los números,
of the country we should go to,
del país debíamos visitar,
really disillusioned with our work,
con nuestro trabajo,
a really good job,
un buen trabajo
who could really tell us were the Iraqis,
eran los iraquíes,
our analysis, let alone question it.
de ver o cuestionar nuestro análisis.
numbers more accurate
be able to question them.
a cuanta gente se pudiera.
and sharing them with the public.
y compartirlas con el público.
you don't have to be a geek or a nerd.
no se necesita ser un nerdo.
they're used by people
la gente las usa
pretendiendo ser modestos.
while pretending they're humble.
some bad statistics.
malas estadísticas.
is: Can you see uncertainty?
¿Se distingue incertidumbre?
people's relationship with numbers,
de la gente con los números,
con las encuestas políticas
with political polls
is actually to report the facts
es informar sobre hechos
can actually damage democracy
pueden dañar la democracia
don't bother to vote for that guy,
ni voten por ese tipo,
about the accuracy of this endeavor.
de la exactitud de esta tarea.
in the UK, Italy, Israel
en el Reino Unido, Italia, Israel
elección presidencial de EE. UU.,
US presidential election,
resultados electorales
to predict hospital admissions.
para predecir admisiones hospitalarias.
from an academic study to draw this.
de un estudio académico para sacarlo.
las encuestas se han vuelto inexactas.
polling has become so inaccurate.
obtener un muestreo representativo
to get a really nice representative sample
their phones to pollsters,
por teléfonos encuestas,
people might lie.
la gente miente.
know that to look at the media.
para mirar los medios.
of a Hillary Clinton win
de que Hillary Clinton ganara
para describir la temperatura.
to describe the temperature.
of 230 million voters in this country
de 230 millones de votantes en este país
will overstate certainty, and it works --
exageran exactitud, y funciona;
our brains to criticism.
nos insensibilizan a la crítica.
you might feel skeptical.
pueden sentirse escépticos.
of objective science,
to better communicate this to people,
de comunicar esto a la gente,
in our numbers.
en los números.
real data sets,
de información real,
hand-drawn visualizations,
hechas a mano,
how imprecise the data is;
cuán imprecisos son los datos;
y la visualizó.
of finding out the probability
en lugar de encontrar la probabilidad
en algún mes,
distribution of flu season.
de la temporada de gripa.
mostrada en febrero.
data visualization,
de datos más responsable,
the exact probabilities,
las probabilidades exactas
people to get their flu jabs
a tomar sus vacunas
these imprecisions,
estas imprecisiones,
walk away with a specific number,
se quedan con un número específico,
leave a huge mark on our lives.
marcan nuestras vidas.
Americans have shorter life expectancies
americanos tienen menos esperanza de vida
prisioneros en EE. UU. en aislamiento
can be kept in solitary confinement cells
of an average parking space.
espacio de aparcamiento promedio.
is also to remind people
es también recordarle a la gente
statistical concepts,
muy importantes,
contains 6.23 fecal accidents."
sufre de 6,23 accidentes fecales".
swimming pool in the country
que cada alberca en el país
which comes from the CDC,
provenientes de la CDC,
instalaciones acuáticas.
redistributing poop.
redistribuyendo excremento.
how misleading averages can be.
cuán engañosos son los promedios.
that you guys should be asking yourselves
que se deben hacer
about averages in a way,
de los promedios,
why people are so frustrated
of who's winning and who's losing
quién va ganando o perdiendo
are frustrated with global averages
se frustra con los promedios globales
with their personal experiences.
con sus experiencias personales.
se relacionan con la vida cotidiana.
data relates to their everyday lives.
called "Dear Mona,"
"Querida Mona,"
with questions and concerns
preguntas y preocupaciones
en camas separadas con mi esposa?",
in a separate bed to my wife?"
of natural causes?"
por causas naturales?".
because they make you think
porque te hacen pensar
y comunicar estos números.
and communicate these numbers.
"How much pee is a lot of pee?"
es demasiada orina?",
that the visualization makes sense
de que la visualización tenga sentido
in the appendix of an academic study.
de un estudio académico.
these numbers on urination volume,
estos números de volumen de orina,
and try it for yourself.
con una botella.
no es necesariamente
has to relate specifically to you.
se relacione con Uds.
were issued fines in France
recibieron multas en Francia
or wear the face veil.
o no uso el velo facial.
is to get as much context as possible.
como sea posible
from one data point,
is five percent,
según el estatus educativo,
by educational status --
que fueran a la universidad--
wanted you to go to college --
de varones es mayor
it was the other way around.
era al revés.
of the biggest changes
once you look beyond the averages.
si ven más allá de los promedios.
you can change the story.
se puede cambiar la historia.
that I want you guys to think about
que quiero que piensen
de cómo se comunican los datos,
data is communicated,
es igual de importante.
matters just as much.
and actually kind of boring,
y un tanto aburrida,
you can take to check this.
para revisar esto.
musulmanes del país apoyaban la yihad,
in this country support jihad,
the original questionnaire.
el cuestionario original.
who reported on that statistic
que reportaron esa estadística
lower down on the survey
en la encuesta
how they defined "jihad."
definir "yihad".
para ser más religioso".
to be more religious."
y violenta contra los no creyentes".
"violent holy war against unbelievers."
it's totally possible
es posible que nadie
who defined it as violent holy war
como guerra santa
cómo se realizó la encuesta.
how the survey was carried out.
de participación libre,
on the internet and completed it.
en internet y responderla.
if those people even identified as Muslim.
se identificaban como musulmanes.
respondents in that poll.
en esa encuesta.
de musulmanes en este país,
Muslims in this country,
one in every 5,000 Muslims
a uno de cada 5000 musulmanes
tres razones por las que
are often better than private statistics.
son mejores que las privadas.
hundred people, maybe a thousand,
cientos, o miles,
productos para el cuidado de la piel,
skin care products in 2005,
para decir que trabajaron.
to claim that they work.
gran interés por obtener bien los números,
interest in getting the numbers right,
los números correctos.
no son así.
they're totally impartial,
their jobs regardless of who's in power.
sin importar quién esté en el poder.
to a couple hundred people.
de los que he hablando
I keep on referencing
de Estadísticas Laborales,
businesses in this country.
negocios en el país.
de una compañía privada,
that comes from a private company,
y pruébenla
and a bunch of friends, test it out,
que los números se equivocan.
you can say the numbers were wrong.
estadísticas gubernamentales?
government statistics?
todo lo que necesitan ver.
on the chart you need to see.
porque si lo hacen,
altogether, because if you do,
políticas públicas ciegamente,
decisions in the dark,
para guiarnos.
interests to guide us.
ABOUT THE SPEAKER
Mona Chalabi - Data journalistMona Chalabi tries to take the numb out of numbers. She's left with lots of "ers."
Why you should listen
After working for a humanitarian organisation, Mona Chalabi saw how important data was, but also how easily it could be used by people with their own specific agendas. Since then, her work for organizations like Transparency International and The Guardian has had one goal: to make sure as many people as possible can find and question the data they need to make informed decisions about their lives.
Chalabi is currently the Data Editor of the Guardian US, where she writes articles, produces documentaries and turns data into illustrations and animations. In 2016, her data illustrations were commended by the Royal Statistical Society.
Mona Chalabi | Speaker | TED.com