Home

ISLAMIC MEDICAL EDUCATION RESOURCES-03

0104-STATISTICS REVISION

By Professor Omar Hasan Kasule sr.

SAMPLES

Target and study populations

Simple random sample

Stratified random sample

Systematic random sample

Multi-stage random sample

Cluster random sample

 

SCALES

Qualitative scales: nominal, ordinal, and ranked

Quantitative scale: numerical discrete and numerical continuous (ratio & interval)

 

HYPOTHESES

Hypothesis testing and the scientific method

The null and the alternative hypotheses

The concept, meaning and significance of the p-value

Rejection/non-rejection of the null hypothesis

Errors of statistical testing

 

DATA COLLECTION, MANAGEMENT, & PRESENTATION

Sources of data: General population & household census, Vital statistics, Routinely-collected data, Epidemiological studies, Special surveys

 

Data collection: Face-to-face questionnaire administration, Questionnaire administration by telephone, Questionnaire administration by mail, Computer-administered questionnaire:

 

Data management: Data coding, entry, storage, & retrieval, Data processing

 

Data presentation: Data grouping, Data tabulation, Diagrams, Shapes of distributions, Misleading diagrams

 

RATES AND PROPORTIONS

Difference between a rate and a proportion

Crude rates, specific rates, and standardized/adjusted rates.

Population pyramid: age and sex-structure of different populations

 

MEASURES OF CENTRAL TENDENCY

Definition, properties, advantages, and disadvantages of various types of averages: the arithmetic mean, mode, & median

Relation between the mean, mode, and median in a normal distribution curve, a curve skewed to the right, and a curve skewed to the left

 

MEASURES OF DISPERSION/VARIATION

Definition, properties, advantages and disadvantages of the variance, standard deviation, and z-score

Definition and use of quartiles and percentiles; relation between percentile, standard deviation, and area under a normal curve

 

INFERENCE ON CONTINUOUS DATA USING MEANS

Inference on 1 mean: t test, z test

Inference on 2 means: t test. Difference between paired and unpaired tests

Inference on 3 or more sample means: F test (ANOVA)

 

INFERENCE ON DISCRETE DATA USING PROPORTIONS

Inference on 1 sample proportion: z test (chi test)

Inference on 2 sample proportions: chi square test (paired and unpaired). Inference on 3 or more sample proportions: chi square test

 

SIMPLE CORRELATION & SIMPLE REGRESSION

Simple Linear Correlation

Simple Linear Regression

Concepts of dependent and independent variables

Linear correlation coefficient (range -1 to +1)

The regression line and regression equation (Y=a + bx)

Interpreting the intercept, a (+ve and -ve)

Interpreting the regression coefficient, b (range - to to +)

Use of the regression line/equation for (a) test of association (b) prediction (interpolation and extrapolation)

 

INTRODUCTION TO EPIDEMIOLOGY

Definition of epidemiology as study of distribution and determinants of disease

Pioneers in epidemiology: John Graunt, William Farr, John Snow, Robert Koch,

 

DISEASE DESCRIPTION and MEASUREMENT

Disease occurrence measures: incidence and prevalence

Epidemicity: epidemic, endemic, pandemic, epidemic curve, slow and acute epidemics, visibility of the epidemic

Determinants of disease: demography, nutritional, infection, genetics, environment, SES, behavior

Prevention: primary, secondary, tertiary, surveillance

Definition and computation of incidence rate, prevalence, and cumulative incidence

 

STUDY DESIGN

Experimental studies, observational studies

Cross-sectional observational study

Case control observational study

Follow-up observational study

 

STUDY ANALYSIS

Tests/measures of association: continuous data (t test, F test, regression); discrete data (chi-square)

Measures of effect: risk ratio, odds ratio, attributable rate

Definition and sources of 3 biases: 1. misclassification bias 2. selection bias 3. confounding bias.

Confounding bias: prevention at design stage (stratification & matching), treatment at analysis stage (standardization, stratified analysis, multivariate adjustment).

 

COMMUNICABLE DISEASES

Attack rates (primary & secondary), case fatality ratio.

The causal triangle: agent, host, and disease

Prevention of communicable disease (primary, secondary, tertiary) and surveillance

Investigation and management of a disease out-break

 

NON-COMMUNICABLE DISEASES

Changing incidence patterns: difference between developed and developing countries, time trends

Description of general risk factors: nutrition, alcohol, tobacco, physical inactivity, environment

Risk factors of and prevention of cancer (lung, breast, cervix, liver,

Omar Hasan Kasule, Sr, Apr 2001