Lecture#1
|
Definition and types of biostatistics and variables.
|
Department: Family and community
medicine
tutorial:
Learning objectives:
By the end of this lecture, students should be
able to:
1. Identify rational of statistics in medical
practice.
2. Recognize the relation between statistics,
epidemiology and computer science ,
complementary discipline in medical practice .
3. Define biostatistics with
its types and the difference between them.
4. define a variable and identify different types
of variables and scales of measurement
5. Differentiate between, a variable, data and
information.
Detailed content:
Definitions:
- Variables, data, information ,statistics; both
descriptive & analytic.
- Types of data;
quantitative & qualitative .
- Scales of
measurement : nominal, ordinal, numerical; interval & ratio.
Methods: Lecture
Readings:
We don't closely follow specific text for this
course. We will provide lecture notes through out the course. however, if u
wish to consult a book, we suggest the following boxes:
1. "An introduction to Statistical Methods
& Data Analysis." R.Lyman Ott & Michael Longnechar, Duxbury
Press, 5th edition , 2001
2."Statistics: The Art and Science of
learning from data". Agresti and Franklin
3."Introduction to the practice of
statistics, 5th edition, Moore & Mcabe.
4. "Text book: Statistical Concepts &
Methods". Bhattacharyya and Johson (wiley)
Lecture#2
|
Descriptive statistics : Data presentation; Tables and graphs
|
Department: Family and community
medicine tutorial:
Learning objectives:
By the end of this lecture, students should be
able to:
1. Define population and sample
Sources of data collection -2.
Identify the different :
-
Data forms , coding table and coding dictionary
-
Data reduction
3. Describe the basic concepts and different
methods of tabular presentation of data
4. Describe the basic concepts and different
methods of graphical presentation of data.
Detailed content:
1. Collection of data:
sources, whole population studies, sampling.
2. Describing data resulting from a categorical
variable: Tally and frequency tables, bar graph, bie chart.
3. Describing data resulting from two categorical variables: contingency tables,
marginal percentages, conditional percentages, side by side bar graphs
4. Summary characteristics and features of
measurements data: Distribution shape, center, variability.
5. Graphical and numerical summaries for
measurements data:
- Univariate analysis: line plots, dot
plots, histograms, stem leaf plots, box plots, line graphs, time plot.
-
Brivariate analysis: Use of scatter plots, correlation, fitting a line.
Methods: Lecture/ Computer lab.
Readings:
We don't closely follow specific text for this
course. We will provide lecture notes through out the course. however, if you
wish to consult a book, we suggest the following boxes:
1. "An introduction to Statistical Methods
& Data Analysis." R.Lyman Ott & Michael Longnechar, Duxbury
Press, 5th edition , 2001
2."Statistics: The Art and Science of
learning from data". Agresti and Franklin
3."Introduction to the practice of
statistics, 5th edition, Moore & Mcabe.
4. "Text book: Statistical Concepts &
Methods". Bhattacharyya and Johnson (wiley).
Teaching location:
Learning objectives:
By the end of this lecture, students should be
able to:
1. Define and determine the various measures of
central tendency
2. Define and determine the various measures of
dispersion
3. Explain why the mean ± 2 standard deviations
is often used to establish the "normal range" and what practical
difficulties might be encountered using this procedure in clinical practice.
Detailed content:
1. Descriptive statistics
for numerical data: measures of central tendency (mean, median, mode).
2. Descriptive statistics
for nominal data: proportions & percentages.
3. Measures of
variability: the range, the variance, the standard deviation and coefficient
of variation.
Methods: Lecture/
Computer lab.
Readings:
We don't closely follow specific text for this
course. We will provide lecture notes through out the course. however, if you
wish to consult a book, we suggest the following boxes:
1. "An introduction to Statistical Methods
& Data Analysis." R.Lyman Ott & Michael Longnechar, Duxbury
Press, 5th edition , 2001
2."Statistics: The Art and Science of
learning from data". Agresti and Franklin
3."Introduction to the practice of
statistics, 5th edition, Moore & McCabe.
4. "Text book: Statistical Concepts & Methods".
Bhattacharyya and Johnson (Wiley).
Lecture# 4
|
Normal distribution and z- score
|
Department: Family and community medicine tutorial:
Teaching location:
Learning objectives:
By the end of this lecture, students should be
able to:
1. Define the normal distribution and describe its characteristics
2. Contrast the features of a normal distribution to those of skewed
distribution or kertotic one
3. Define the standard normal distribution and how to use it to
estimate the standard value (z- value).
4. Define and estimate the standard error and to know its applications
in the clinical practical field.
Detailed content:
1- Normal distribution curve and percentiles.
2- Skew ness and kurtosis.
3- z-score
4- Standard error of the mean (SEM).
5- Standard error of proportion
Methods: Lecture/
Computer lab.
Readings:
We don't closely follow specific text for
this course. We will provide lecture notes through out the course. however,
if you wish to consult a book, we
suggest the following boxes:
1. "An introduction to Statistical Methods
& Data Analysis." R.Lyman Ott & Michael Longnechar, Duxbury
Press, 5th edition , 2001
2."Statistics: The Art and Science of
learning from data". Agresti and Franklin
3."Introduction to the practice of
statistics, 5th edition, Moore & Mcabe.
4. "Text book: Statistical Concepts &
Methods". Bhattacharyya and Johson (wiley)
Lecture# 5
|
Interval estimation "confidence
internals"
|
Department: Family and community
medicine
tutorial:
Teaching location:
Learning objectives:
By the end of this lecture,
students should be able to:
1.Use the standard error to
estimate the different confidence limits( range of probability):
- Estimate
confidence interval for a mean
- Estimate
confidence interval for a proportion
2. Interpret statements
containing confidence limits.
3. Make defensible inferences
about samples and populations:
- Estimate a single
population mean μ when the standard deviation σ is
Known.
- Estimate a single
population proportion.
Detailed content:
Point estimate , interval estimate, the upper and lower confidence
limits ( range of probability ) i.e. confidence intervals; 90% C.I, 95% C.I,
99% C.I .
Methods:
Lecture.
Readings:
We don't closely follow specific
text for this course. We will provide lecture notes through out the course.
however, if u wish to consult a book, we suggest the following boxes:
1. "An introduction to
Statistical Methods & Data Analysis.". R.Lyman Ott & Michael
Longnechar, Duxbury Press, 5th edition , 2001
2."Statistics: The Art and
Science of learning from data". Agresti and Franklin
3."Introduction to the
practice of statistics, 5th edition
,Moore & Mcabe.
4. "Text book: Statistical
Concepts & Methods". Bhattacharyya and Johson (wiley)
Lecture#6
|
Statistical inference
|
Department: Family and community medicine
Tutorial:
Teaching location:
Learning objectives:
By the end of this lecture, students should be able
to:
1. Recognize the
difference between population distribution and sampling distribution.
2. Outline basic concepts
of hypothesis testing.
3. Recognize the types of
question-problems (description, estimation, comparison and relation).
4. Interpret statements
of statistical significance with regard to comparisons of means and
frequencies, and explain what is meant by a statement such as
"p<0.05".
5. Distinguish between
the statistical significance of a result and its importance in clinical
application.
Detailed content:
Null hypothesis, Alternative hypothesis, α error
(type I error), β error (type II error), confidence level, power of the test,
p-value.
Methods: Lecture.
Readings:
We don't closely follow specific text for this
course. We will provide lecture notes through out the course. however, if you
wish to consult a book, we suggest the following boxes:
1. "An introduction to Statistical Methods
& Data Analysis". R.Lyman Ott & Michael Longnechar, Duxbury
Press, 5th edition , 2001
2."Statistics: The Art and Science of
learning from data". Agresti and Franklin
3."Introduction to the practice of
statistics". 5th edition, Moore & Mcabe.
4. "Text book: Statistical Concepts &
Methods". Bhattacharyya and Johson (wiley)
Lecture#7
|
Comparison between
means: "t" tests
|
Department: Family and community
medicine
Tutorial:
Teaching location:
Learning objectives:
By the end of this lecture,
students should be able to:
1. Describe meaning and uses of
hypothesis testing of arithmetic means.
2. Recognize and calculate the
statistical test appropriate for comparison between one
sample mean & a population mean.
3. Recognize and calculate the
statistical test appropriate for comparison between two
independent means.
4. Recognize and calculate the
statistical test appropriate for comparison between two
means in a paired sample.
5. Recognize the statistical
test appropriate for comparison between more than two
independent means.
Detailed content:
One
sample "t" test, Student "t" test, Paired "t"
test, ANOVA test.
Methods:
Lecture/ Computer lab.
Readings:
We don't closely follow specific
text for this course. We will provide lecture notes through out the course.
however, if you wish to consult a book, we suggest the following boxes:
1. "An introduction to
Statistical Methods & Data Analysis." R.Lyman Ott & Michael
Longnechar, Duxbury Press, 5th edition , 2001
2."Statistics: The Art and
Science of learning from data". Agresti and Franklin
3."Introduction to the
practice of statistics, 5th edition, Moore & Mcabe.
4. "Text book: Statistical
Concepts & Methods". Bhattacharyya and Johson (wiley) .
Lecture#8
|
Comparison between proportions:
z-test and x2 test
|
Department: Family and community medicine
Tutorial:
Teaching location:
Learning objectives:
By the end of this lecture, students should be
able to:
- Recognize and calculate the statistical test appropriate for
comparison between two independent proportions.
- Recognize and calculate the statistical test appropriate for
comparison between more than two independent proportions
- Recognize and calculate the statistical test appropriate for
comparison between two proportions in a paired sample .
Detailed content:
"z" test, x2 text,
Mc Nemar test, Fisher exact test.
Methods: Lecture/
Computer lab.
Readings:
We don't closely follow specific text for this
course. We will provide lecture notes through out the course. however, if you
wish to consult a book, we suggest the following boxes:
1. "An introduction to Statistical Methods
& Data Analysis." R.Lyman Ott & Michael Longnechar, Duxbury
Press, 5th edition , 2001
2."Statistics: The Art and Science of
learning from data". Agresti and Franklin
3."Introduction to the practice of
statistics, 5th edition, Moore & Mcabe.
4. "Text book: Statistical Concepts &
Methods". Bhattacharyya and Johson (wiley) .
Lecture#9
Sampling techniques :
probability & non probability
Department: Family and community
medicine
Tutorial:
Teaching location:
Learning objectives:
By the end of this lecture,
students should be able to:
1. List at least four advantages
of sampling.
2. Describe different sampling
techniques.
3. Explain sampling bias and
describe how random sampling operates to avoid bias in the of data
collection.
4. List types and methods of
probability sampling.
5. List types and methods of
non- probability sampling.
6. Distinguish between standard
deviation and standard error, and give one example of the use of each.
Detailed content:
Steps in
sample selection: sampling frame, sample size, randomization.
Simple
random sample, systemic random sample, stratified random sample, cluster
sample, multistage random sample.
Convenience
sample, Consecutive sample, Judgmental sample.
Methods:
Lecture/ Computer lab.
Readings:
We don't closely follow specific
text for this course. We will provide lecture notes through out the course.
however, if u wish to consult a book, we suggest the following boxes:
1. "An introduction to
Statistical Methods & Data Analysis." R.Lyman Ott & Michael
Longnechar, Duxbury Press, 5th edition , 2001
2."Statistics: The Art and
Science of learning from data". Agresti and Franklin
3."Introduction to the
practice of statistics, 5th edition, Moore & Mcabe.
4. "Text book: Statistical
Concepts & Methods". Bhattacharyya and Johson (wiley).
Lecture#10
Correlation:
Department: Family and community
medicine
Tutorial:
Teaching location:
Learning objectives:
By the end of this lecture,
students should be able to:
- Recognize the question problem of correlation
between two interdependent variables.
- Interpret the relationship between two
variables as displayed on a scatter gram.
- Distinguish between positive, negative and zero
correlation.
- Recognize and interpret Pearson's correlation
for two variables of interval.
- Detect the significance of correlation between
two interdependent variables.
Detailed content:
Scatter gram,
correlation coefficient "r", and t-test of the "r".
Readings:
We don't closely follow specific
text for this course. We will provide lecture notes through out the course.
however, if you wish to consult a book, we suggest the following boxes:
1. "An introduction to
Statistical Methods & Data Analysis." R.Lyman Ott & Michael
Longnechar, Duxbury Press, 5th edition , 2001
2."Statistics: The Art and
Science of learning from data". Agresti and Franklin
3."Introduction to the
practice of statistics, 5th edition, Moore & Mcabe.
4. "Text book: Statistical
Concepts & Methods". Bhattacharyya and Johson (wiley) .
Methods:
Lecture/ Computer lab.
Lecture#11
Regression analysis:
Department: Family and community
medicine
Tutorial:
Teaching location:
Learning objectives:
By the end of this lecture,
students should be able to:
1. Develop
a predictive model to predict a dependent variable value given a specific
level of an independent variable which is correlated to it.
2. Explain
the information provided by a regression equation.
3. Distinguish
between a simple regression equation and a multiple regression one.
Detailed content:
1.
The best fitted line in a scatter plot, its slope and intercept.
2.
Residual error
3.
Linear regression equation
4.
Regression coefficient.
5.
Coefficient of determination.
6. Multiple regression equation.
Methods:
Lecture/ Computer lab.
-Objectives of the lectures
Biostatistics:
1- Develop the concept of critical
interpretation of data.
2- Enable future doctors to use
statistical principles to improve
their professional work.
3- Help future researchers in the
medical field to make optimum use
of their resources to reach valid
research conclusions.
4- Help students to understand medical
literatures and make critical
Learning
Resources
Textbooks:
Statistics In
Medicine.
Thedore Colton,Sc.D.
N.B. Some lectures will have handouts to clear and simplify
what is become in these textbooks. However, the department have elaborated
these handouts to the point where they represent a data textual description
of the material within the course curriculum
and so can be used as the primary reference in lieu of the textbook
e.g. biostatistics lectures.
Example of
Questions with model answer
Select the best answer from the following multiple
answers: Two tests (ELIZA & PCR) were conducted for diagnosis of HBV
among 600 persons. Results showed that a total of 300 persons were + ve by
ELIZA screening test, and 50 % of these screening test + ve were also
positive by PCR confirmatory test. On the other hand, 90 % of those were
screening test –ve are also –ve by PCR diagnostic test.
Answer the following questions: (you can use approximation)
-The sensitivity of ELIZA test was:
a.
30 %.
b.
50 %.
c.
64 %.
d.
83 %.√
e.
90 %.
-The specificity of ELIZA test was:
a.
30 %.
b.
50 %.
c.
64 %.√
d.
80%.
e.
90 %.
-The positive value of predicted test
(positive predicted test) of ELIZA was:
a.
30 %.
b.
50 %. √
c.
64 %.
d.
80%.
e.
90 %.
|