ESTADÍSTICA DESCRIPTIVA

• ¿QUÉ ES LA ESTADÍSTICA?
• ¿QUÉ ES LA CIENCIA?
• ¿QUÉ ES EL CONOCIMIENTO?
• ¿QUÉ ES LA INFORMACIÓN?
• ¿QUÉ ES LA REALIDAD?
• ¿QUÉ RELACIÓN EXISTE ENTRE TODOS ESTOS
CONCEPTOS?
Para que un fenómeno pueda ser objeto de
estudio de la ciencia tiene que ser medible

????????

Dra. Ma Fernanda Castilla P.

 Variables

Cada uno de los atributos de un fenómeno que

pueden ser medidos.

Dra. Ma Fernanda Castilla P.

 Medición

Asignar una categoría o valor a una variable

mediante una instrumento y una escala

Dra. Ma Fernanda Castilla P.

 Escala

Conjunto de todos los valores o categorías

que puede tomar una variable.

Dra. Ma Fernanda Castilla P.

Instrumento de medición
Cuestionarios

Escalas Clínicas

Dra. Ma Fernanda Castilla P.

 Tipos de variables

• Cuantitativas/ dimensionales/ de Talla, edad, concentración de

razón-De intervalo una substancia en sangre
– Continuas Número de hijos, años
– Discretas cumplidos

• Cualitativas/ Nominales/
Categóricas Sí/No Presente/ausente
- Dicotómicas
Religión, lugar de nacimiento
- Policotómicas

• Ordinales/
Intensidad de un soplo, dolor,
Cuasidimensionales depresión,
Dra. Ma Fernanda Castilla P.
• Frecuencia cardiaca
• Peso al nacer Tipo de variable

• Religión
• Apgar Instrumento de medición
• Macrosomía
• Estadio de Tanner
Escala
• Coeficiente intelectual
• Numero de crisis epilépticas en un mes
• Control del Asma
• Felicidad

Dra. Ma Fernanda Castilla P.

 Transformación de variables
Cuantitativa Cualitativa Ordinal
Tensión arterial Mm Hg Hipertenso/ Normal/Prehiperte
Normotenso nsión/ Hipertensión
Estatus Ingreso mensual Marginado/No Bajo/Medio/Alto
socioeconómico marginado
Inteligencia Retraso Mental/ No Coeficiente
retraso mental intelectual
Gluocosa en ayuno Mg/dl Control/ Descontrol Menor de 100/100-
126/ mayor de 126

Otras transformaciones:
-Logarítimica
-Raiz cuadrada
-Elevar al cuadrado
--Otras
Según su papel en el estudio de investigación

• Variable Supuesta

independiente CAUSA

Supuesto EFECTO

• Variable
dependiente
Int J Prev Med. 2013 February; 4(2): 173–179.

Determinants of Stunting in School-Aged

Children of Tehran, Iran
Fatemeh Esfarjani,1,2 Roshanak Roustaee,1 Fatemeh Mohammadi,1 and Ahmad Esmaillzadeh3,4

… 3147 school children were selected by

multistage cluster random sampling method from

5 districts of Tehran. Anthropometric

measurements were done and stunting was

defined as height for age less than the 5th

percentile of CDC2000 cut-off points. Eighty six

Rev Invest Clin. 1995 Nov-Dec;47(6):433-8.
 Gestational diabetes in a Mexican-U.S. border population:
prevalence and epidemiology.
Meza E, Barraza L, Martínez G, Fernández V, Ramos-Jáquez E,
Cano-Vargas C, Valdez-Torres A, Izaguirre R.
OBJECTIVE:
To determine the prevalence of gestational diabetes (GD) in a
Mexican-U.S. border population and to observe the response of such a
population to a 100 gram challenge of glucose.
RESEARCH DESIGN AND METHODS:
The sample was a cross section of 519 pregnant women accesing
different health institutions in Juarez, Mexico, with 24-36 weeks of
gestation. An oral glucose tolerance test (OGTT) with a 100 gram
challenge of glucose was done on all. Results were interpreted
Dra. Ma Fernanda Castilla P.
Pediatrics. 2013 Jan;131(1):e96-102. doi: 10.1542/peds.2012-1008. Epub 2012 Dec 24.
Validation of a migraine interview for children and adolescents.
Lateef T, Cui L, Heaton L, Nakamura EF, Ding J, Ahmed S, Merikangas KR.

OBJECTIVE: To date there are no structured interviews to ascertain the diagnostic

criteria for headache in children. The objective of this study was to assess the validity of

the Diagnostic Interview of Headache Syndromes-Child Version (DIHS-C), which was

developed at the National Institute of Mental Health for a community-based family

study of headache syndromes and comorbid disorders.

METHODS: The DIHS-C is a fully structured diagnostic interview composed of an open-

ended clinical history, modules with key symptoms for each of the major headache

subtypes, and associated impairment, duration, frequency, course, and treatment. This

article presents the validation of the interview in a sample of 104 children evaluated as

part of a community-based family study of migraine.

Dra. Ma Fernanda Castilla P.
RESULTS: The sensitivity of interview diagnosis compared with an expert neurologist's
Pediatrics. 2013 Mar;131(3):e770-8. doi: 10.1542/peds.2012-1907. Epub 2013 Feb 25.
Maternal alcohol use and sudden infant death syndrome and
infant mortality excluding SIDS.
O'Leary CM, Jacoby PJ, Bartu A, D'Antoine H, Bower C.

This study investigates the association between maternal alcohol-

use disorder and sudden infant death syndrome (SIDS) and infant
mortality not classified as SIDS using linked, population-based health
and mortality data.
Exposed mothers were identified through the presence of an
International Classification of Diseases 9/10 alcohol diagnosis, a
proxy for alcohol-use disorder, recorded on health, mental health,
and/or drug and alcohol datasets (1983-2005). Comparison mothers
without an alcohol diagnosis were frequency matched to exposed
mothers on maternal age within maternal race and year of birth of
their children. All offspring with their birth recorded on the
Midwives Notification System compose the exposed (n = 21 841)
and comparison (n = 56 054)Dra.cohorts.
Ma Fernanda Castilla P.
Cases of SIDS (n = 303) and
N Engl J Med. 2013 Jan 31;368(5):425-35. doi: 10.1056/NEJMoa1202851.

Antibiotics as part of the management of

severe acute malnutrition.
Trehan I, Goldbach HS, LaGrone LN, Meuli GJ, Wang RJ, Maleta KM, Manary MJ.

METHODS:

……we randomly assigned Malawian children, 6 to 59 months of age, with severe

acute malnutrition to receive amoxicillin, cefdinir, or placebo for 7 days in

addition to ready-to-use therapeutic food for the outpatient treatment of

uncomplicated severe acute malnutrition. The primary outcomes were the rate

of nutritional recovery and the mortality rate.

Dra. Ma Fernanda Castilla P.
ESTADÍSTICA ESTADÍSTICA
DESCRIPRIVA INFERENCIAL
• Extraer conclusiones acerca
• Resumir los datos de la “realidad”(población
de una muestra total) a partir de los datos de
una muestra.

• Calcular intervalos de
confianza de una
estimación

• Probar una hipótesis y

calcular la probabilidad de
error (tipo I y II)
 ESTADÍSTICA DESCRIPTIVA:
RESUMIR DATOS
VARIABLES CUALITATITAVAS

– Frecuencias
• Frecuencia absoluta

• Frecuencia relativa (proporción):

» Prevalencia
» Incidencia
 Ejercicios
• Clasificiación por propósito

• Variables
– Dependiente-Independiete
– Tipo, Instrumento de medición y escala

• Pregunta de investigación (PICO, PEO o PETS)

• Clasificación del estudio por arquitectura

– Descriptivo-Comparativo
– Observacional-Experimental
– Longitudinal-Transversal
– Causa-Efecto/ Efecto-causa
– Retrolectivo-Prolectivo

Dra. Ma Fernanda Castilla P.

Int J Prev Med. 2013 February; 4(2): 173–179.

Determinants of Stunting in School-Aged Children of Tehran, Iran

Fatemeh Esfarjani,1,2 Roshanak Roustaee,1 Fatemeh Mohammadi,1 and Ahmad Esmaillzadeh3,4

… 3147 school children were selected by multistage cluster random

sampling method from 5 districts of Tehran. Anthropometric

measurements were done and stunting was defined as height for age less

than the 5th percentile of CDC2000 cut-off points. Eighty six stunted

children were identified and considered as case group. After matching for

age, sex and residence area, 308 non-stunted children were randomly

selected as control group. Required data were collected by trained

nutritionists using questionnaires.

Rev Invest Clin. 1995 Nov-Dec;47(6):433-8.
 Gestationaldiabetes in a Mexican-U.S. border population:
prevalence and epidemiology.
Meza E, Barraza L, Martínez G, Fernández V, Ramos-Jáquez E, Cano-Vargas C, Valdez-Torres A,
Izaguirre R.

OBJECTIVE:
To determine the prevalence of gestational diabetes (GD) in a Mexican-U.S.
border population and to observe the response of such a population to a 100
gram challenge of glucose.
RESEARCH DESIGN AND METHODS:
The sample was a cross section of 519 pregnant women accesing different
health institutions in Juarez, Mexico, with 24-36 weeks of gestation. An oral
glucose tolerance test (OGTT) with a 100 gram challenge of glucose was done
on all. Results were interpreted according to the diagnostic criteria of
O'Sullivan and Mahan. Information on family history of diabetes, obstetric
Dra. Ma Fernanda Castilla P.
history, height and weight was obtained.
Pediatrics. 2013 Jan;131(1):e96-102. doi: 10.1542/peds.2012-1008. Epub 2012 Dec 24.

Validation of a migraine interview for children and adolescents.

Lateef T, Cui L, Heaton L, Nakamura EF, Ding J, Ahmed S, Merikangas KR.

OBJECTIVE: To date there are no structured interviews to ascertain the diagnostic criteria for headache in

children. The objective of this study was to assess the validity of the Diagnostic Interview of Headache

Syndromes-Child Version (DIHS-C), which was developed at the National Institute of Mental Health for a

community-based family study of headache syndromes and comorbid disorders.

METHODS: The DIHS-C is a fully structured diagnostic interview composed of an open-ended clinical

history, modules with key symptoms for each of the major headache subtypes, and associated

impairment, duration, frequency, course, and treatment. This article presents the validation of the

interview in a sample of 104 children evaluated as part of a community-based family study of migraine.

RESULTS: The sensitivity of interview diagnosis compared with an expert neurologist's diagnosis of

migraine was 98%, and the specificity was 61%. Similar levels of sensitivity and specificity were found by

gender and age of the children.

Dra. Ma Fernanda Castilla P.

Pediatrics. 2013 Mar;131(3):e770-8. doi: 10.1542/peds.2012-1907. Epub 2013 Feb 25.
Maternal alcohol use and sudden infant death syndrome and infant
mortality excluding SIDS.
O'Leary CM, Jacoby PJ, Bartu A, D'Antoine H, Bower C.

This study investigates the association between maternal alcohol-use

disorder and sudden infant death syndrome (SIDS) and infant mortality not
classified as SIDS using linked, population-based health and mortality data.

Exposed mothers were identified through the presence of an International

Classification of Diseases 9/10 alcohol diagnosis, a proxy for alcohol-use
disorder, recorded on health, mental health, and/or drug and alcohol
datasets (1983-2005). Comparison mothers without an alcohol diagnosis
were frequency matched to exposed mothers on maternal age within
maternal race and year of birth of their children. All offspring with their
birth recorded on the Midwives Notification System compose the exposed
(n = 21 841) and comparison (n = 56 054) cohorts. Cases of SIDS (n = 303)
and infant mortality excluding SIDS (n = 598) were identified through linkage
with the Western Australian Mortality Register. Analyses were conducted by
using Cox regression and results presented as adjusted hazard ratios (aHRs)
and 95% confidence intervalsDra.(CIs).
Ma Fernanda Castilla P.
N Engl J Med. 2013 Jan 31;368(5):425-35. doi: 10.1056/NEJMoa1202851.

Antibiotics as part of the management of

severe acute malnutrition.
Trehan I, Goldbach HS, LaGrone LN, Meuli GJ, Wang RJ, Maleta KM, Manary MJ.

METHODS:

……we randomly assigned Malawian children, 6 to 59 months of age, with severe

acute malnutrition to receive amoxicillin, cefdinir, or placebo for 7 days in addition to

ready-to-use therapeutic food for the outpatient treatment of uncomplicated severe

acute malnutrition. The primary outcomes were the rate of nutritional recovery and

the mortality rate.

Dra. Ma Fernanda Castilla P.
Representación gráfica de distribución de los
datos

“Una imagen dice más que mil palabras”

 Datos categóricos (cualitativos)

• Gráfica de sectores ( gráfica de pie)

Vìa de nacimeinto

Porcentaje Porcentaje
Frecuencia Porcentaje válido acumulado
Válidos parto 16 25.4 26.7 26.7
cesàrea 44 69.8 73.3 100.0
Total 60 95.2 100.0
Perdidos Sistema 3 4.8
Total 63 100.0

El total suma 100% !!

 Datos categóricos u ordinales
ESCOLARIDADA

Porcentaje Porcentaje
Frecuencia Porcentaje v álido acumulado
Válidos 0 1 1.6 1.6 1.6
1 1 1.6 1.6 3.2
2 5 7.9 8.1 11.3
3 29 46.0 46.8 58.1
4 17 27.0 27.4 85.5
5 3 4.8 4.8 90.3
6 6 9.5 9.7 100.0
Total 62 98.4 100.0
Perdidos Sistema 1 1.6
Total 63 100.0
 Datos cuantitativos
• Tallo y hojas
Peso_inic Stem-and-Leaf Plot

Frequency Stem & Leaf

4.00 Extremes (=<0)

6.00 4 . 288999
13.00 5 . 0233444578888
18.00 6 . 001122344455667788
13.00 7 . 0001234555568
4.00 8 . 0068
1.00 9. 0
1.00 Extremes (>=104)

Stem width: 10
Each leaf: 1 case(s)
 Datos cuantitativos
• Histograma Polígono de frecuencias
 Datos cuantitativos
• Caja y bigotes
MEDIDAS DE TENDENCIA CENTRAL
• Media aritmética: suma de todos los valores
entre el número de mediciones (promedio)

• Mediana: el valor que divide a la población en

dos grupos del mismo tamaño (el valor de en
medio)

• Moda: el valor más frecuente

 Medidas de Dispersión
• Desviación absoluta: promedio de la distancia (absoluta,
sin signo) de cada valor a la media

• Varianza (s2, 2): promedio de los cuadrados de la

distancia de cada valor a la media

• Desviación estándar/ desviación típica (s, ): Raiz

cuadrada de la varianza
 Medidas de posición
• Cuantiles
– Terciles: dividen el grupo en 3 partes iguales
– Cuartiles: dividen el gurpo en 5
– Quintiles…

– Percentiles: dividen el grupo en 100 partes

iguales
 Descriptivos

Estadí st ico Error típ.

Peso_inic Media 60.58 2.588
Interv alo de conf ianza Límite inf erior 55.40
para la media al 95% Límite superior
65.75

Media recortada al 5% 62.08

Mediana 63.50
Varianza 401.821
Desv . tí p. 20.045
Mínimo 0
Máximo 104
Rango 104
Amplitud intercuartil 18
Asimetrí a -1.559 .309
Curtosis 3.829 .608

Percentil es

Percentiles
5 10 25 50 75 90 95
Promedio Peso_inic
.00 48.00 54.00 63.50 71.88 79.80 87.90
ponderado(def inición 1)
Bisagras de Tukey Peso_inic 54.00 63.50 71.75
 ESTADÍSITICA DESCRIPTIVA
Medidas de Medidas de Medida estadística
tendencia dispersión más usada
central
Dimensionales / Media, Desviación estándar, Media
Cuantitativas Mediana, Moda Intervalo mínimo-
máximo, cuantiles
Ordinales Mediana, Moda Intervalo mínimo- Mediana
máximo, cuantiles
Nominales/Categó Moda Conjunto de Frecuencias
ricas categorías

Dra. Ma Fernanda Castilla P.