Statistical

Methods
By
Prof. Dr/ Monir Hussein Bahgat
Professor of Internal Medicine, Mansoura University.
Trainer of SPSS and statistical analysis skills, Mansoura University.
Quality manager, Specialized Medical Hospital, Mansoura University.
Acknowledgments

This two-year curriculum was developed through a participatory and collaborative approach
between the Academic faculty staff affiliated to Egyptian Universities as Alexandria University, Ain
Shams University, Cairo University, Mansoura University, Al-Azhar University, Tanta University, Beni
Souef University, Port Said University, Suez Canal University and MTI University and the Ministry of
Health and Population(General Directorate of Technical Health Education (THE). The design of this
course draws on rich discussions through workshops. The outcome of the workshop was course
specification with Indented learning outcomes and the course contents, which served as a guide to
the initial design.

We would like to thank Prof.Sabah Al- Sharkawi the General Coordinator of General Directorate
of Technical Health Education, Dr. Azza Dosoky the Head of Central Administration of HR
Development, Dr. Seada Farghly the General Director of THE and all share persons working at
General Administration of the THE for their time and critical feedback during the development of this
course.

Special thanks to the Minister of Health and Population Dr. Hala Zayed and Former Minister
of Health Dr. Ahmed Emad Edin Rady for their decision to recognize and professionalize health
education by issuing a decree to develop and strengthen the technical health education curriculum
for pre-service training within the technical health
Contents

Acknowledgments ................................................................. ii

Course Description ............................................................... .7

Preface .............................................................................. 9

Chapter 1: Introduction to statistical methods .............................. 12

Chapter 2: Proper selection of statistical methods .......................... 23

Chapter 3: Measures of central tendency (Measures of location) ....... 32

Chapter 4: Measures of dispersion (Spread) ................................. 42

Chapter 5: Measures of frequency ............................................ 54

Chapter 6: Hypothesis testing .................................................. 60

Chapter 7: Estimation statistics ............................................... 78

Appendix: A ....................................................................... 88

Appendix: B ....................................................................... 91

Appendix: C ....................................................................... 95

Appendix: D ....................................................................... 99

References ........................................................................ 100

 Statistical Methods

Course Description

This course is prepared for undergraduate students aiming at introducing the student
both to statistical reasoning and to the most commonly used statistical techniques. It
should also support the students who master this material to be able to select,
implement, and interpret the most common types of analyses as they undertake
research in their own disciplines.

Core Knowledge

On successful completion of this course, the student will be able to :

1- Define statistical methods.
2- Identify the two types of statistical methods; descriptive statistics and inferential
statistics.
3- Recognize the commonly used terms in statistical methods.
4- Select the appropriate statistical method according to the goal and type of data.
5- Describe the different methods of descriptive statistics including measures of
central tendency and measures of dispersion.
6- Determine the proper use of inferential statistics including its two main types;
hypothesis testing and estimation statistics.

4
 Statistical Methods

Core General Skills

On successful completion of this course, the student will be able to:

1. Able to search for a computer software or website to help perform statistical
analysis in easier way.
2. Work hardly to improve the skills in applying statistical methods.
3. Demonstrate caution and proficiency in applying statistical methods.

Core Professional Skills

On successful completion of this course, the student will be able to:

1. Sample a population in a representative manner (particularly simple random
sampling).
2. Analyze the raw data in a proper way.
3. Label values correctly.
4. Make hypothesis testing.
5. Solve statistical problems.
6. Criticize a research paper.

Teaching and learning methods

1- Lectures using power point presentations.

2- Positive interaction with the lecturer by asking questions or answering them.
3- Practical sessions to solve statistical exercises.
4- Hand-outs to simplify the scientific material.
5- External readings of specialized books.
6- Training to answer model question exercises.

Teaching and learning methods for students with disabilities

1- Remedial lectures.
2- Remedial practicing lessons.
3- Simplified sketches for different statistical methods.
4- Training to answer model question exercises.

5
 Statistical Methods

Core Skills

on successful completion of this course, the student will be able to :

1. Solve exercises on the commonly used statistical methods.
2. Apply the proper statistical method to a given dataset.
3. Interpret the results of statistical methods.
4. Report the results of statistical methods.
5. Compare the different types of statistical methods.
6. Hypothesize in a correct way what to be tested.

‫توصٌف مقرر دراسى‬

‫بٌانات المقرر‬

ً‫ شعبة تسجٌل طب‬/ ‫الثانٌة‬: ‫المستوى‬/ ‫الفرقة‬ Statistical Methods ‫ طرق إحصائٌة‬:‫اسم المقرر‬ : ‫الرمز الكودى‬

‫ عملى‬21 ‫ نظري‬3 :‫عدد الوحدات الدراسٌة‬

: ‫التخصص‬

‫تقوٌم الطالب‬

1- Practice test. ‫األسالٌب المستخدمة‬

2- Mid-term test evaluation.
3- Final written test.

‫التوقٌت‬
Final theoretical written exam: 3-hours.

1- Mid-tem test = 20 marks

2- Written test 100 marks ‫توزٌع الدرجات‬
3- Practical test = 80 marks
4- Total marks = 200

6
 Statistical Methods

‫قائمة الكتب الدراسٌة والمراجع‬

Handouts for the lectures and practical sections ‫مذكرات‬

Statistical methods: 1st edition. ‫كتب ملزمة‬

Freud RJ, Wilson WJ and Mohr DL (2010). Statistical Methods
(Third Edition). Amsterdam, Boston, Heidelberg, London New
York, Oxford, Paris, San Diego San Francisco, Singapore,
Sydney, Tokyo. Elsevier Inc. ‫كتب مقترحة‬

The Journal of Modern Applied Statistical Methods:

https://digitalcommons.wayne.edu/jmasm/ ...... ‫دورٌات علمٌة أو نشرات‬
‫الخ‬

7
 Statistical Methods

Course Overview

ID Theory Practice
Introduction: Define and classify
statistical methods. Outline
1st week Plot a flow chart for statistical methods
importance of studying statistical
methods.

Essential terminology to
The student is asked to create explanatory
2nd week understand common statistical
example for each statistical term.
terms.

Proper selection of the suitable

The student is offered different scenarios and
statistical method for a particular
3rd week then is asked to nominate the appropriate
goal and a particular type of
statistical method to each scenario.
variable.

4th week Revision for part I Revision for part I

5th week Descriptive statistics: Measures of Problem solving for mean, median and mode
central Tendency.

Descriptive statistics: Measures of Problem solving for ranges and standard

6th week
dispersion. deviation.

Descriptive statistics: Measures of Problem solving for rate, ratio, proportion, and
7th week
frequencies. percentage.

8th week Revision for part II Revision for part II

The student is asked to create explanatory

9th week Introduction to hypothesis testing
examples

5-step procedure for hypothesis The student is asked to perform the procedure
10th week
testing working on an explanatory example.

The student is asked to perform the procedure

11th week Hypothesis testing for a
proportion working on an explanatory example.

Estimation statistics: Confidence The student is asked to perform the procedure

12th week
Interval working on an explanatory example.

13th week Revision for part III Revision for part III

8
 Statistical Methods

Preface

Statistical methods are mathematical formulas, models, and techniques that

are used in statistical analysis of raw research data.

The application of statistical methods extracts information from research data

and provides different ways to assess the strength of research outputs.

This book is designed for undergraduate students aiming at introducing the

student both to statistical reasoning and to the most commonly used statistical
techniques.

The goal is that students who master this material will be able to select,
implement, and interpret the most common types of analyses as they undertake
research in their own disciplines.

Moreover, they should be able to read research articles and in most cases
understand the descriptions of the statistical results and how the authors used
them to reach their conclusions.

They should understand the pitfalls of collecting statistical data, and the roles
played by the various mathematical assumptions.

Statistics can be studied at several levels:

9
 Statistical Methods

On one hand, students can learn by repetition how to plug numbers into formulas,
or more often now, into a computer program, and draw a number with a neat
circle around it as the answer. This limited approach is mind distressing, and
rarely leads to the kind of understanding that allows students to critically select
methods and interpret results.

On the other hand, there are numerous textbooks that provide introductions to
the elegant mathematical backgrounds of the methods. Although this is a much
deeper understanding than the first approach, its prerequisite mathematical
understanding closes it to practitioners from many other disciplines.

This book seems to take a middle way by presenting enough of the formulas to
motivate the techniques, and by illustrating their numerical application in a small
example.

However, the focus of the discussion is on the selection of the technique, the
interpretation of the results, and a critique of the validity of the analysis.

The student is advised to focus on these skills.

This book is divided into three parts:

 Part I (Introduction to statistical methods).

 Part II (Methods of descriptive statistics).
 Part III (Methods of inferential statistics).

10
 Statistical Methods

Flow Chart of Statistical Methods:

Statistical
Methods

Descriptive Inferential
Statistics Statistics

Measures of
Measures of Hypothesis Estimation
Central
Dispersion Testing Statistics
Tendency

11
 Statistical Methods

Introduction to Statistical Methods

Introduction to statistical methods

Part I (Introduction to statistical methods):

I.1. Essential terminology:

- Population and sample:

Definition:

A population is a data set representing the entire entity of interest while a sample
is a data set consisting of a portion of a population. Normally a sample is obtained
in such a way as to be representative of the population.

Explanatory example:

You are interested in studying prevalence of systemic hypertension among

Mansoura University students. The study population will be „all‟ students in the
university. As this will consume time and resources, you‟ll carefully select a
representative sample to do the study so that the results obtained from this
sample can be generalized to the population from which this sample was selected.

12
 Statistical Methods

- Variables:

Definition:

Variable is a characteristic of some event, object, or person that can take on

more than one value.

Explanatory example:

When a study involves the two types of sex (coded as 1 if male and 2 if female),
sex in this study is considered a variable. In another study performed on males
only, sex is constant and is not considered as a variable.

Variable types:

Variables are classified into qualitative (categorical) and quantitative.

Qualitative variable is a variable that expresses a qualitative attribute (has no

measurement unit) while quantitative variable has a measurement unit.

Qualitative variables are of two types: Nominal (categories have no meaningful

order) and Ordinal (categories have meaningful order). Dichotomous variable is a
sub-type of categorical variable with only two categories.

Quantitative variables are of two types: Continuous (data are continuous points on
the scale and therefore include fractions) and Discrete (data are discrete points
on the scale and therefore always come in integer form).

Quantitative variables are of two measurement types: Interval variable (a

measurement where the difference between two values is meaningful) and a
ratio variable (has all the properties of an interval variable, and also has a clear

13
 Statistical Methods

definition of zero. When the variable equals 0.0, there is none of that variable).
When working with ratio variables, but not interval variables, you can look at the
ratio of two measurements.

Explanatory examples:

Sex (coded as 1 if male and 2 if female) is a nominal variable because it has no

measurement unit with no meaningful order of the codes

Education level (coded as 1 if illiterate or just read & write, 2 if educated to a

level below university and 3 if educated at a university level or more) is an ordinal
variable because it has no measurement unit with meaningful order of the codes
as 3 means higher education level than 2 and 2 means higher education level than
1.

Age (measured in years) is a continuous variable because it has a measurement

unit (years) and fractions are accepted (patient‟s age might be 23.5 years).

Number of children for each female in a study is a discrete variable because it has
a measurement unit (children number; e.g., 2 children, 3 children, and so on) and
fractions are not accepted (a female can‟t have 3.5 children).

Temperature (expressed in F or C) is an example of interval variable. The

difference between a temperature of 100 degrees and 90 degrees is the same
difference as between 90 degrees and 80 degrees. A temperature of 0.0 on either
of those scales does not mean 'no heat'.

Weight is a ratio variable. Therefore, a weight of 60 kg is twice a weight of 30 kg,

while a temperature of 40 degrees C is not twice as hot as 20 degrees C, because
temperature C is not a ratio variable like weight.

14
 Statistical Methods

- Data set:

Definition:

A set of data is a collection of observed values representing one or more

characteristics of some objects or units.

Explanatory example:

You asked a 11 patients about their age (in years), sex (coded as 1 if male and 2 if
female) and education level (coded as 1 if illiterate or just read & write, 2 if
educated to a level below university and 3 if educated at a university level or
more). You tabulated the results as follows:

Table (1): How data set looks like

ID Age Sex Education level

1 21 1 1

2 24 1 2

3 23 2 2

4 25 2 3

5 26 1 1

6 30 2 2

7 31 1 3

8 32 1 3

9 20 2 1

10 21 2 1

11 24 2 2

15
 Statistical Methods

This data set is a collection of observed values representing age (continuous

variable), sex (nominal variable) and education level (ordinal variable) of 11
patients.

- Descriptive and inferential statistics:

Definition:

Descriptive statistics intend to describe a data set with summary charts and
tables, but do not attempt to draw conclusions about the population from which
the sample was taken. You are simply summarizing the data you have like telling
someone the key points of a book (executive summary) as opposed to just handing
them a thick book (raw data).

Conversely, with inferential statistics, you are testing a hypothesis and drawing
conclusions about a population, based on your sample.

Explanatory example:

Descriptive statistics:

Let‟s say you‟ve tested 50 university students for hepatitis C antibody. You have a
bunch of data plugged into your spreadsheet and now it is time to share the
results with someone. You could hand over the spreadsheet and say “here‟s what

16
 Statistical Methods

I learned” (not very informative), or you could summarize the data with some
charts and graphs that describe the data and communicate some conclusions (e.g.
10% of university students have positive anti-HCV antibody). This would sure be
easier for someone to interpret than a big spreadsheet. There are hundreds of
ways to visualize data, including data tables, pie charts, line charts, etc. That is
the idea of descriptive statistics. Note that the analysis is limited to your data
and that you are not extrapolating any conclusions about the full population (the
whole university students).

Inferential statistics:

Let‟s continue with HCV example. Let‟s say you wanted to know the prevalence
in Mansoura University students. Well, there are >110,000 students and it would
be impossible to test every single student for anti-HCV. Instead, you would try to
test a representative sample of students and then extrapolate your sample results
to the entire population. While this process is not perfect and it is very difficult
to avoid errors, it allows researchers to make well-reasoned inferences about the
population in question. This is the idea behind inferential statistics. Getting a
representative sample is really important. There are many methods of sampling
strategies, including random sampling. A true random sample means that
everyone in the target population has an equal chance of being selected for the
sample. Another key component of proper sampling is the sample size. Obviously,
the larger the sample size, the better, but there are trade-offs in time and money
when it comes to obtaining a large sample.

17
 Statistical Methods

There are online calculators as well as software (free and commercial) that help
determine appropriate sample sizes.

Examples of online calculators include:

https://www.surveysystem.com/sscalc.htm

https://www.calculator.net/sample-size-calculator.html

http://clincalc.com/stats/samplesize.aspx

Examples of software include:

https://www.cdc.gov/epiinfo/index.html (free)

http://www.gpower.hhu.de/en.html (free)

https://www.ncss.com/software/pass/ (commercial)

When it comes to inferential statistics, there are generally two forms: estimation
statistics and hypothesis testing.

Estimation Statistics:

“Estimation statistics” is a way of saying that you are estimating population values
based on your sample data. Let‟s think back to our sample HCV data. First, let‟s
assume that we had a true random sample of 50 students from this university and
that our full target population is >110,000 students. Let‟s say that 10% of
students in our sample had positive anti-HCV. Can we safely extrapolate that 10%
of all university students also will have positive anti-HCV? Is that the true value of
the university? Well, we can‟t say with 100% confidence, but–using inferential

18
 Statistical Methods

statistical techniques such as the “confidence interval” we can provide a range of

students that test positive for anti-HCV with some level of confidence.

Hypothesis Testing

Hypothesis testing is simply another way of drawing conclusions about a

population parameter (“parameter” is simply a number, such as a mean, that
includes the full population and not just a sample).

With hypothesis testing, one uses a test such as T-Test, Chi-Square, or ANOVA to
test whether a hypothesis about the mean is true or not. Again, the point is that
this is an inferential statistic method to reach conclusions about a population,
based on a sample set of data.

19
 Statistical Methods

Practical exercises

1. Serum albumin measured in g/dl is

a. A nominal variable.
b. An ordinal variable.
c. An interval variable.
d. A ratio variable.
2. Eye color is:
a. A nominal variable.
b. An ordinal variable.
c. An interval variable.
d. A ratio variable.
3. You examined the quality of chest x-rays whether poor, fair, good or
excellent. This is:
a. A nominal variable.
b. An ordinal variable.
c. An interval variable.
d. A ratio variable.
4. A measurable characteristic of a population is:
a. A parameter
b. A statistic
c. A sample
d. An experiment

20
 Statistical Methods

5. A measurable characteristic of a sample is:

a. A parameter
b. A statistic
c. A population
d. An experiment
6. A subset of a population is:
a. a parameter
b. a statistic
c. a sample
d. an experiment
7. the body weight in kilograms is
a. A categorical (nominal) variable.
b. An ordinal variable.
c. A continuous variable.
d. A discrete variable.

8. A true random sample means that:

a. Everyone in the target population has an equal chance of being

selected for the sample.

b. Everyone in the target population has a 75% chance of being selected

for the sample.

c. Everyone in the target population has a 25% chance of being selected

for the sample.

d. Not everyone in the target population has an equal chance of being

selected for the sample.

21
 Statistical Methods

Answers:

1. D.
2. A.
3. B.
4. A.
5. B.
6. C.
7. C.
8. A.

22
 Statistical Methods

Proper selection of Statistical Methods

Proper selection of the statistical method

I.2. Proper selection of the statistical method:

Statistical methods are mathematical formulas, models, and techniques that are
used in statistical analysis of raw research data. The application of statistical
methods extracts information from research data and provides different ways to
assess the strength of research outputs.

To properly select which statistical method to use, we need to know our aim
(goal) and the type of data (or variable) to be tested. The goal may be just
descriptive or it might be performing a comparison or looking for a relationship.
Variables may be nominal (including the dichotomous variable), ordinal, or
quantitative. For quantitative variables, we need to know if this variable is
normally distributed (Gaussian distribution) or not (non-Gaussian distribution).
This can be done by many statistical methods including Kolmogorov-Smirnov test
(for large data sets) and Shapiro-Wilk test (for small data sets <50) where data
will be considered normally distributed if the test result is insignificant (p value >
0.050) and will be considered non-normally distributed (skewed) if the test result
is significant (p value ≤ 0.050).

23
 Statistical Methods

Figure 2.1: Normal (Gaussian) distribution:

µ (pronounced mu) = population mean,  (pronounced sigma) = population

standard deviation

The normal distribution is bell shaped and symmetric about the mean.

The interval (µ ± 1) contains approximately 68% of the observations.

The interval (µ ± 2) contains approximately 95% of the observations.

The interval (µ ± 3) contains virtually all of the observations (99.7%).

The following table simplify the choice of the proper statistical method:

24
 Statistical Methods

Table (1): Choosing a proper statistical method for analysis

Type of data / variable

- Quantitative
Aim (goal) Quantitative
(skewed distribution) Nominal
(normal distribution)
- or ordinal

Describe a group Mean ± SD Median (IQR) Frequency (%)

Compare one-group to a One-sample Chi-

One-sample t-test Wilcoxon test
hypothetical value Square test

Compare paired data Paired-samples t-

Wilcoxon test McNemar test
(two sets) test

Compare repeated Repeated-measures

Friedman test Cochrane Q test
measures (>two sets) ANOVA

Independent-
Compare two-groups Mann-Whitney test Chi-Square test
samples t-test

Compare > two groups One-Way ANOVA test Kruskal Wallis H test Chi-Square test

Relationships between variables:

Contingency
Association Pearson‟s correlation Spearman‟s correlation
coefficient

Logistic
Prediction Linear regression e.g., ordinal regression
regression

SD=Standard deviation, IQR=Interquartile range, ANOVA=Analysis of variance.

25
 Statistical Methods

Part II (Methods of descriptive statistics):

The two most important aspects of describing a data set are the location (central
tendency) and the dispersion (spread) of the data.

In other words, we need to find a number that indicates where the observations
are on the measurement scale and another to indicate how widely the
observations vary.

26
 Statistical Methods

Practical exercises

1. You collected the age in years for a sample of 40 patients. To test for
normality, you will use:
a. Kolmogorov-Smirnov test.
b. Shapiro-Wilk test.
c. Mann-Whitney test.
d. ANOVA test.

2. You tested the normality of age in years of a sample of 300 patients using
Kolmogorov-Smirnov test and you found that p value = 0.475. this means
that the age distribution is:
a. Normal.
b. Non-Gaussian.
c. Positively-skewed.
d. Negatively-skewed.

3. The mean age ± standard deviation of a sample of 100 patients equals 50 ±

5. Considering that age is normally distributed, you are expected that nearly
68 patients will have their age:
a. Exactly 50 years.
b. Between 45 and 50 years.
c. Between 45 and 55 years.
d. Between 40 and 60 years.

27
 Statistical Methods

4. To describe sex variable of a group of patients, you will use.

a. Mean and standard deviation.
b. Median and interquartile range.
c. Proportions.
d. None of the above.

5. You measured the sleeping hours for 50 patients first after taking a placebo
and then after taking a sleeping pill that is claimed to be better than
placebo. Sleeping hours were normally distributed after both placebo and
sleeping pill. To test this hypothesis, you will use:
a. One-sample t-test.
b. Paired-samples t-test.
c. Independent-samples t-test.
d. ANOVA test.

6. You measured the sleeping hours for a group of 50 patients after taking a
placebo and for another group of 50 patients after taking a sleeping pill that
is claimed to be better than placebo. The two groups are of similar age and
sex (matched groups). Sleeping hours were normally distributed in both
groups. To test this hypothesis, you will use:
a. One-sample t-test.
b. Paired-samples t-test.
c. Independent-samples t-test.
d. ANOVA test.

28
 Statistical Methods

7. You measured the sleeping hours after taking one of three sleeping pills to
compare their effect. Each pill was tested in a separate group of 50
patients; and the three groups were of similar age and sex (matched
groups). Sleeping hours were normally distributed in each of the three
groups. To test this hypothesis, you will use:
a. One-sample t-test.
b. Paired-samples t-test.
c. Independent-samples t-test.
d. ANOVA test.

8. You want to compare the proportion of fall out of bed in a medical

department (50 patients) against a surgical department (50 patients). You
found that 25 patients fell in medical department (50%) and 10 patients fell
in surgical department (20%). To compare the fall rate between the two
department, you will use:
a. Paired-samples t-test.
b. Independent-samples t-test.
c. ANOVA test.
d. Chi-square test.

29
 Statistical Methods

9. You want to test for a possible linear association between the heart rate
(measured in beats / minute) of a 100 healthy women and their age
(measured in years). You assumed that as age goes up, heart rate will go
down (negative correlation). You found that both heart rate and age were
normally distributed. To test this hypothesis, you will use:
a. Pearson‟s correlation.
b. Spearman‟s correlation.
c. Contingency coefficient.
d. Chi-square test.

10. You want to test if it is possible to predict the heart rate of a healthy
woman (measured in beats / minute) by knowing her age (measured in
years). To test this hypothesis, you will use:
a. Pearson‟s correlation.
b. Spearman‟s correlation.
c. Linear regression.
d. Logistic regression.

11. You want to test if it is possible to predict that a patient admitted to

hospital will fall out of bed (i.e., fall or no fall). The predictors (independent
variables) are age of the patient (measured in years) and the admission
department (medical versus surgical). To test this hypothesis, you will use:
a. Pearson‟s correlation.
b. Spearman‟s correlation.
c. Linear regression.
d. Logistic regression.

30
 Statistical Methods

Answers:

1. B.
2. A.
3. C.
4. C.
5. B.
6. C.
7. D.
8. D.
9. A.
10. C.
11. D.

31
 Statistical Methods

Measures of central tendency (measures of location)

Measures of central tendency (measures of location)

Three different measures are available to describe data location including:

- Mean
- Median
- Mode

Mean:

The most frequently used measure of location is the arithmetic mean,

usually referred to simply as the mean or the average.

Definition:

The mean is the sum of all the observed values divided by the number of
values.

32
 Statistical Methods

Formula:

n = Sample size

yi = observed values of the variable Y (where i = 1, 2, . . , n).

 = the sum of.

Explanatory example:

For the example in table (1) for the 15 patients, their mean age will be
26.07 years:

Mean = (21+24+23+25+26+30+31+32+20+21+24+26+28+29+31)/15

[Try to calculate it yourself].

Median:

This is another useful measure of location.

Definition:

The median of a set of observed values is defined to be the middle value

when the measurements are arranged from lowest to highest; that is, 50% of
the measurements lie above it and 50% fall below it.

The precise definition of the median depends on whether the number of

observations is odd or even as follows:

33
 Statistical Methods

- If n is odd, the median is the middle observation; hence, exactly (n -

1)/2 values are greater than and (n - 1)/2 values are less than the
median, respectively.
- If n is even, there are two middle values and the median is the mean
of the two middle values and n/2 values are greater than and n/2
values are less than the median, respectively.

Explanatory example:

a. „n‟ is odd number of observations:

To calculate the median for the following dataset: 2, 1, 3, 5, 3, 6, 4.

1. Arrange from lowest to highest: 1, 2, 3, 3, 4, 5, 6

2. As the n is odd (7), median is the middle value (4th value) which is 3

b. „n‟ is even number of observations:

To calculate the median for the following dataset: 2, 1, 3, 5, 3, 6, 4, 2.

1. Arrange from lowest to highest: 1, 2, 2, 3, 3, 4, 5, 6

2. As the n is even (8), median is the mean of the two middle values
(mean of 4th and 5th values) which are 3 and 3. So, median = (3+3)/2
= 6/2 =3

Comparing mean and median:

Which one should be used for descriptive statistical method?

The choice of the measure to be used may depend on its ultimate

interpretation and use.

34
 Statistical Methods

For symmetric or nearly symmetric distributions (bell-shaped), the mean

and median will be the same or nearly the same, while for skewed
distributions the value of the mean will tend to be “pulled” toward the long
tail.

This is illustrated in the following figure:

Figure 3.1: Normal and skewed distribution:

This phenomenon can be explained by the fact that the mean can be
interpreted as the center of gravity of the distribution. That is, if the
observations are viewed as weights placed on a plane, then the mean is the
position at which the weights on each side balance. It is a well-known fact
of physics that weights placed further from the center of gravity exert a
larger degree of influence (also called leverage); hence the mean must shift
toward those weights in order to achieve balance. However, the median
assigns equal weights to all observations regardless of their actual values;
hence, the extreme values have no special leverage.

35
 Statistical Methods

Explanatory example:

For the following two datasets X and Y we‟ll calculate mean and median for
each dataset:

X: 1, 2, 3, 3, 4, 5

Y: 1, 1, 1, 2, 5, 8

Mean values:

For X = (1 + 2 + 3 + 3 + 4 + 5) / 6 = 18 / 6 = 3

For Y = (1 + 1 + 1 + 2 + 5 + 8) / 6 = 18 / 6 = 3

So, the mean value for these two different datasets is equal.

Median values:

For X = (3 + 3) / 2 = 3

For Y = (1 + 2) / 2 = 1.5

So the median value for these two datasets is different.

The reason for this difference is due to different distribution which is

symmetric for variable X (note that mean and median for this dataset is the
same) and skewed to the right for variable Y.

36
 Statistical Methods

Pearl:

“Use the mean as the single measure of location unless the distribution of
the variable is skewed.”

Table (1)

Normal distribution Skewed distribution

Mean Median

The mean is calculated using the value of each observation, so all the
information available from the data is utilized. This is not so for the
median. For the median, we only need to know where the “middle” of the
data is. Therefore, the mean is the more useful measure and, in most cases,
the mean will give a better measure of the location of the data. However,
as we have seen, the value of the mean is heavily influenced by extreme
values and tends to become a distorted measure of location for a highly
skewed distribution. In this case, the median may be more appropriate.

Mode:

Definition:

The mode is the most frequently occurring value.

This measure may not be unique in that two (or more) values may occur
with the same greatest frequency.

37
 Statistical Methods

Also, the mode may not be defined if all values occur only once, which
usually happens with continuous numeric variables.

Explanatory example:

For the following dataset:

1, 1, 2, 1, 3, 2, 4, 3, 1, 3, 4, 2, 3, 1, 1, 1, 1, 3, 4, 5, 6, 3, 6, 2, 5, 1, 1

The most frequently occurring value is „1‟, which occurs 10 times. So the
mode is „1‟.

38
 Statistical Methods

Practical exercises

1. Which one of the following is a measure of central tendency?

a. Range.
b. Interquartile range.
c. Standard deviation.
d. Median.

2. For a dataset of 25, 18, 5, 12, 24, and 16, the mean is
a. 12
b. 16.7
c. 20.5
d. 18

3. For a dataset of 25, 18, 5, 12, 24, and 16, the median is
a. 16
b. 17
c. 18
d. 20

4. If 5, 3, 4 and 2 of your sample cases have age of 20, 23, 30, and 25
years, respectively, the age mode is:
a. 20 years.
b. 23 years.

39
 Statistical Methods

c. 25 years.
d. 30 years.

5. The median is a better measure of central tendency than the mean if:
a. The variable is discrete.
b. The distribution is skewed.
c. The variable is continuous.
d. The distribution is symmetric.

6. The median can be defined as:

a. The 25th percentile.
b. The 50th percentile.
c. The 75th percentile.
d. The average.

7. For a dataset of 1, 4, 6, 3, 2, and 5:

a. The mean is more than the median.
b. The mean less than the median.
c. The mean equals the median.
d. The mean does not equal the median.

40
 Statistical Methods

Answers:

1. D.
2. B.
3. B.
4. A.
5. B.
6. B.
7. C.

41
 Statistical Methods

Measures of dispersion (spread)

Measures of dispersion (spread):

Although location is generally considered to be the most important single

characteristic of a distribution, the variability or dispersion of the values is
also very important.

The mean or median of a variable provides an inadequate description of the

distribution of that variable since the list of values would include a wide
range.

Therefore, description is best presented by using a measure of central

tendency as well as a measure of dispersion.

For normally distributed data, using mean and standard deviation is the
most appropriate while for non-normally distributed data, using median and
interquartile range is more appropriate.

Measures of dispersion include:

- Range.
- Variance.
- Standard deviation.

42
 Statistical Methods

Range:

Definition:

The difference between the largest and smallest observed values.

Although it is the simplest and most obvious measure of variability, the

range has one very serious drawback: It completely ignores any information
from all the other values in the data.

Explanatory example:

For a dataset of 2,4,5,4,6,3,7,5,1,9,5

The minimum value is 1

The maximum value is 9

So, the range equals 9-1 = 8

Interquartile Range (IQR):

Definition:

The interquartile range is the length of the interval between the 25th
and 75th percentiles and describes the range of the middle half of the
distribution.

Formula:

IQR = Upper quartile (Q3) – Lower quartile (Q1)

43
 Statistical Methods

Steps to calculate:

- Organize dataset in ascending order.

- Divide data into two halves:
o Odd number: Midpoint will be the exact middle number.
o Even number: Midpoint will be between the two middlemost
numbers.
- Find the median of:
o Upper half of data (=Q3).
o Lower half of data (=Q1).
- Subtract Q3 – Q1 to determine the IQR.

Explanatory example:

- For a dataset of 7,6,3,4,8,2,1,5

o Data in ascending order: 1,2,3,4,5,6,7,8
o Data in two halves: 1,2,3,4 and 5,6,7,8
o Q3 = (6 + 7) / 2 = 13 / 2 = 6.5
o Q1 = (2 + 3) / 2 = 5 / 2 = 2.5
o IQR = 6.5 – 2.5 = 4
- For a dataset of 9,7,6,3,4,8,2,1,5
o Data in ascending order: 1,2,3,4,5,6,7,8,9
o Data in two halves: 1,2,3,4 and 6,7,8,9 (5=exact middle number)
o Q3 = (7 + 8) / 2 = 15 / 2 = 7.5
o Q1 = (2 + 3) / 2 = 5 / 2 = 2.5
o IQR = 7.5 – 2.5 = 5

44
 Statistical Methods

Variance:

Definition:

The variance is mathematically defined as the average of the squared

differences from the mean.

Formula:

The sample variance, denoted by s2 (pronounced sigma square), of a set of

n observed values having a mean y¯ is the sum of the squared deviations
divided by n - 1:

n = Sample size.

yi = observed values of the variable Y (where i = 1, 2, . . , n).

y¯ = mean value.

 = the sum of.

(yi - y¯)2 = Squared deviations.

45
 Statistical Methods

Explanatory example:

For a dataset of 1, 2, 3, 4, 5

To calculate the variance:

- Calculate the mean:

o Mean = (1 + 2 + 3 + 4 + 5) / 5 = 15 / 5 = 3

- Calculate the sum of squared deviations:

o Sum of squared deviations:
= (1-3)2 + (2-3)2 + (3-3)2 + (4-3)2 + (5-3)2
= (-2)2 + (-1)2 + (0)2 + (+1)2 + (+2)2
=4+1+0+1+4
= 10

- Calculate variance:
o Variance:
= Sum of squared deviations / (n - 1)
= 10 / (5 - 1)
= 10 / 4
= 2.5

46
 Statistical Methods

Notes:

1. Note that the variance is actually an average or mean of the squared

deviations and is often referred to as a mean square.

2. Note also that we have divided the sum by (n - 1) rather than n:

- One of the uses of the sample variance is to estimate the population

variance.

- Dividing by n tends to underestimate the population variance;

therefore by dividing by (n - 1) we get, on average, a more accurate
estimate.

- Recall that we have already noted that the sum of deviations (yi - y¯)
= 0; hence, if we know the values of any (n - 1) of these values, the
last one must have that value that causes the sum of all deviations to
be zero.

- Thus there are only (n - 1) “free” deviations. Therefore, the quantity

(n - 1) is called the degrees of freedom.

47
 Statistical Methods

Standard deviation:

Definition:

The standard deviation of a set of observed values is defined to be the

positive square root of the variance.

Formula:

This measure is denoted by s and does have a very useful interpretation as a

measure of dispersion:

s = √𝒗𝒂𝒓𝒊𝒂𝒏𝒄𝒆

Explanatory example:

For the above dataset with variance calculated to be 2.5

s=√

s = 1.58

Usefulness of the Mean and Standard Deviation:

Although the mean and standard deviation (or variance) are only two
descriptive measures, together the two actually provide a great deal of
information about the distribution of an observed set of values.

48
 Statistical Methods

This is illustrated by the empirical rule:

If the shape of the distribution is nearly bell-shaped, the following

statements hold:

- The interval (y¯ ± s) contains approximately 68% of the observations.

- The interval (y¯ ± 2s) contains approximately 95% of the observations.
- The interval (y¯ ± 3s) contains virtually all of the observations.

Note that for each of these intervals the mean is used to describe the
location and the standard deviation is used to describe the dispersion of a
given portion of the data.

Explanatory example:

For a normally distributed dataset, n = 50, mean (y¯) = 20, s = 2

According to empirical rule:

o (y¯ ± s), which is 20 ± 2, defines the interval 18 to 22 and

should include (0.68)*(50) = 34 observations,

o (y¯ ± 2s), which is 20 ± 4, defines the interval from 16 to 24

and should include (0.95)*(50) = 48 observations,

o (y¯ ± 3s) which is 20 ± 6, defines the interval from 14 to 26 and should

include all 50 observations.

49
 Statistical Methods

Calculating the standard deviation (s) from the range:

The empirical rule provides us with a quick method of estimating the

standard deviation of a bell-shaped distribution.

Since at least 95% of the observations fall within 2 standard deviations of

the mean in either direction, the range of the data covers about 4 standard
deviations.

Thus, we can roughly estimate the standard deviation by taking the range
divided by 4.

Explanatory example:

For a dataset, the minimum is 16 and the maximum is example 28,

therefore, the range of this dataset is 28 - 16 = 12. Divided by 4 we get 3.
The estimated standard deviation had a crude value of 3.

50
 Statistical Methods

Practical exercises

1. Which one of the following is a measure of dispersion?

a. Mean.
b. Mode.
c. Standard deviation.
d. Median.
2. For a dataset of 10, 14, 16, 22, 12, and 19, the range is
a. 10.
b. 11.
c. 12.
d. 13.
3. For a dataset of 1, 2, 3, 4, and 5, the variance is:
a. 2
b. 2.5
c. 3
d. 3.5
4. If the variance of a dataset is 4, the standard deviation is
a. 2
b. 4
c. 8
d. 16

51
 Statistical Methods

5. If the variance of a dataset of 100 observations is 4, the standard error

of the mean (SEM) is
a. 0.2
b. 0.4
c. 0.6
d. 0.8
6. The standard error of the mean (SEM) is calculated by dividing the
standard deviation by:
a. Mean.
b. Median.
c. Sample size.
d. Square root of sample size.
7. For a dataset of 0, 1, 2, 3, 4, 5, 6, and 7, interquartile range is:
a. 2
b. 4
c. 6
d. 8
8. In a secondary school, you found that the smallest height among
students was 152 cm and the tallest student has a height of 176 cm. the
estimated standard deviation would roughly be equal to:
a. 2
b. 4
c. 6

52
 Statistical Methods

d. 8

Answers:

1. C.
2. C.
3. B.
4. A.
5. A.
6. D.
7. B.
8. C (Range = 176 – 152 = 24, standard deviation = Range / 4 = 24 / 4 = 6).

53
 Statistical Methods

Measures of frequency:
These measures show how often a value occurs

Measures of frequency include:

 Frequency.
 Ratio.
 Rate.
 Proportion.
 Percentage.

Absolute and relative frequency:

Definition:

Absolute frequency: The number of times a certain value occurs in the data.

Relative frequency: The number of times a certain value occurs in the data
(absolute frequency) relative to the total number of values for that
variable.

The relative frequency may be expressed in ratios, rates, proportions, and

percentages.

54
 Statistical Methods

Ratios:

Definition:

Ratios compare the frequency of one value for a variable with another value
for the same variable.

Explanatory examples:

Example (1):

The sex frequency of a certain disease was studied in 40 participants. They

were found to be 30 females and 10 males. Therefore, the ratio of female to
male is 30:10 (= 3:1).

Example (2):

In 30 participants, a certain adverse effect to an experimental drug

occurred in 2 participants only, therefore, the ratio of an experimental
drug's adverse effect to no adverse effects is 2:28 (= 1:14).

Rate:

Definition:

Rate is the measurement of one value for a variable in relation to the entire
sample of values within a given period.

Explanatory example:

In a total of 30 participants, there are 2 who show adverse effects after

taking an experimental drug; therefore, the rate of adverse effects is 2/30
participants (= 1 / 15).

55
 Statistical Methods

Proportion:

Definition:

Proportion is the fraction of a total sample that has some value.

Explanatory example:

In a total of 30 participants, with two participants having adverse drug

effects, the proportion of adverse effects is 2/30 = 0.067

Percentage:

Definition:

Percentage is another way of expressing a proportion as fraction of 100. The

total percentage of an entire dataset should always add up to 100%.

Explanatory example:

Example (1):

The sex frequency of a certain disease was studied in 40 participants. They

were found to be 30 females and 10 males. Therefore, the percentage of
female is (30 / 40)*100 = (3/4)*100 = (0.75)*100 = 75%. On the other hand,
the percentage of male is (10 / 40)*100 = (1/4)*100 = (0.25)*100 = 25%. Note
that 75% + 25% = 100%.

Example (2):

In total of 30 participants, where 2 experience adverse drug effects,

(2/30)*100 = (0.067)*100 = 6.7% of participants experience adverse effects.

56
 Statistical Methods

Graphical presentation:

The above measures of frequency are often expressed visually in the form of
tables, histograms (for quantitative variables), or pie or bar graphs (for
qualitative variables) to make the information more easily interpretable.

57
 Statistical Methods

Practical exercises

1. Which of the following is a measure of frequency:

a. Rate.
b. Mean.
c. Median.
d. Mode.
2. The number of times a particular value occurs in the data is called:
a. Rate.
b. Absolute frequency.
c. Relative frequency.
d. Proportion.
3. The fraction of a total sample that has some value is called:
a. Rate.
b. Ratio.
c. Proportion.
d. Percentage.
4. In a study on 50 cardiac patients receiving nitrite, severe headache
occurred in 10 patients. The proportion of severe headache is:
a. 10:50
b. 10:40
c. 20%
d. 0.2

58
 Statistical Methods

Answers:

1. A.
2. B.
3. C.
4. D.

59
 Statistical Methods

Hypothesis testing
Hypothesis testing

To reach conclusions about a population, based on a sample dataset using

hypothesis testing, one uses a statistical test such as T-Test, Chi-Square, or
ANOVA test to assess whether a hypothesis about the mean is true or not.

A hypothesis usually results from assumption concerning observed behavior,

natural phenomena, or established theory. If the hypothesis is stated in
terms of population parameters such as the mean and variance, the
hypothesis is called a statistical hypothesis. Data from a sample (which may
be an experiment) are used to test the validity of the hypothesis. A
procedure that enables us to agree or disagree with the statistical
hypothesis using data from a sample is called a test of the hypothesis.

Explanatory example:

A test of the effect of a diet pill on weight loss would be based on observed
weight losses of a sample of healthy adults. If the test concludes the pill is
effective, the manufacturer can safely advertise to that effect.

60
 Statistical Methods

The Hypotheses:

Statistical hypothesis testing starts by making a set of two statements about

the parameter(s) in question.

These are usually in the form of simple mathematical relationships involving

the parameters.

The two statements are exclusive and comprehensive, which means that one
or the other statement must be true, but they cannot both be true.

The first statement is called the null hypothesis and is denoted by H 0, and
the second is called the alternative hypothesis and is denoted by H 1.

Definitions:

The null hypothesis:

The null hypothesis is a statement about the values of one or more

parameters. This hypothesis represents the status quo and is usually not
rejected unless the sample results strongly imply that it is false.

The alternative hypothesis:

The alternative hypothesis is a statement that contradicts the null

hypothesis. This hypothesis is accepted if the null hypothesis is rejected.
The alternative hypothesis is often called the research hypothesis.

Explanatory example:

61
 Statistical Methods

Suppose that you want to conduct a study to test whether smoking causes
lung cancer. The current status at that time is that there is no evidence yet
to say that and this actually what motivated you to conduct this study. So,
the null hypothesis will be that smoking in NOT causing lung cancer
(statement of “no effect”) while the alternative hypothesis will be that
smoking causes lung cancer. You will not be able to reject the null
hypothesis (and accept the alternative hypothesis) until you finished your
research and found that the null hypothesis is false.

The rejection region:

The rejection region (also called the critical region) is the range of values of
a sample statistic that will lead to rejection of the null hypothesis.

Possible Errors in Hypothesis Testing:

The results of a hypothesis test may be subject to two distinctly different

errors, which are called type I and type II errors.

Definition:

Type I error:

A type I error occurs when we incorrectly reject H 0, when H0 is actually true

and our sample-based inference procedure rejects it.

Type II error:

A type II error occurs when we incorrectly fail to reject H 0, that is, when H0
is actually false, and our inference procedure fails to detect this fact.

62
 Statistical Methods

Table (1): Results of a hypothesis test:

H0 in the population
Decision
True False

Ho is rejected Type I error Correct decision

Ho is not rejected Correct decision Type II error

Probabilities of making errors:

Definition:

a- Denotes the probability of making a type I error.

b- Denotes the probability of making a type II error.

Importance:

The ability to provide these probabilities is a key element in statistical

inference, because they measure the reliability of our decisions.

Five-Step Procedure for Hypothesis Testing:

 Step 1:

Specify H0, H1, and an acceptable level of α.

This value is based on the seriousness or cost of making a type I error in the
problem being considered.

63
 Statistical Methods

The significance level of a hypothesis test is the maximum acceptable

probability of rejecting a true null hypothesis.

Why Do We Focus on the Type I Error?

In general, the null hypothesis is usually constructed to be that of the status

quo; that is, it is the hypothesis requiring no action to be taken, no money
to be spent, or in general nothing changed. This is the reason for denoting
this as the null or nothing hypothesis. Since it is usually expensive to
incorrectly reject the status quo than it is to do the reverse, this
characterization of the null hypothesis does indeed cause the type I error to
be of greater concern.

On the other hand, the alternative hypothesis describes conditions for which
something will be done. It is the action or research hypothesis. In an
experimental or research setting, the alternative hypothesis is that an
established (status quo) hypothesis is to be replaced with a new one. Thus,
the research hypothesis is the one we actually want to support, which is
accomplished by rejecting the null hypothesis with a sufficiently low level of
α such that it is unlikely that the new hypothesis will be erroneously
pronounced as true. The significance level represents a standard of
evidence. The smaller the value of α, the stronger the evidence needed to
establish H1.

Historically and traditionally, α has been chosen to have values of 0.10,

0.05, or 0.01, with 0.05 being most frequently used.

 Step 2:

64
 Statistical Methods

Define a sample-based test statistic & rejection region for the specified H0.

The test statistic is a sample statistic whose sampling distribution can be

specified for both the null and alternative hypothesis case.

After specifying the appropriate significance level of α, the sampling

distribution of this statistic is used to define the rejection region.

The rejection region comprises the values of the test statistic for which

- The probability when the null hypothesis is true is less than or equal to
the specified α and
- The probabilities when H1 is true are greater than they are under H 0.
 Step 3:

Collect the sample data and calculate the test statistic.

 Step 4:

Make a decision to either reject or fail to reject H 0.

This decision will normally result in a recommendation for action.

 Step 5:

Interpret the results in the language of the problem in such a way that the
results be usable by the practitioner.

Since H1 is of primary interest, this conclusion should be stated in terms of

whether there was or was not evidence for the alternative hypothesis.

The p value:

65
 Statistical Methods

The p value is the probability of committing a type I error if the actual

sample value of the statistic is used as the boundary of the rejection region.

It is therefore the smallest level of significance for which we would reject

the null hypothesis with that sample.

Consequently, the p value is often called the “attained” or the “observed”

significance level.

It is also interpreted as an indicator of the weight of evidence against the

null hypothesis.

The Probability of a Type II Error:

In presenting the procedures for hypothesis and significance tests we have

concentrated exclusively on the control over α, the probability of making
the type I error.

However, just because that error is the more serious one, we cannot
completely ignore the type II error. There are many reasons for ascertaining
the probability of that error, for example:

- The probability of making a type II error may be so large that the test
may not be useful.
- Because of the trade-off between α and β, we may find that we may
need to increase α in order to have a reasonable value for β.

Power of a test:

Definition:

66
 Statistical Methods

The power of a test is the probability of correctly rejecting the null

hypothesis when it is false. The power of a test is (1 - β) and depends on the
true value of the parameter μ.

As a practical matter we are usually more interested in the probability of

not making a type II error, that is, the probability of correctly rejecting the
null hypothesis when it is false.

Obviously high power is a desirable property of a test. If a choice of tests is

available, the test with the largest power should be chosen. In certain
cases, theory leads us to a test that has the largest possible power for any
specified alternative hypothesis, sample size, and level of significance. Such
a test is considered to be the best possible test for the hypothesis and is
called a “uniformly most powerful” test.

One-tailed and two-tailed hypothesis tests:

Two-tailed test:

If you are using a significance level of 0.05, a two-tailed test allots half of
your alpha to testing the statistical significance in one direction and half of
your alpha to testing statistical significance in the other direction.

This means that .025 is in each tail of the distribution of your test statistic.

When using a two-tailed test, regardless of the direction of the relationship

you hypothesize, you are testing for the possibility of the relationship in
both directions.

67
 Statistical Methods

Figure 6.1: Two-tailed test

One-tailed test:

If you are using a significance level of .05, a one-tailed test allots all of your
alpha to testing the statistical significance in the one direction of interest.

This means that .05 is in one tail of the distribution of your test statistic.

When using a one-tailed test, you are testing for the possibility of the
relationship in one direction and completely disregarding the possibility of a
relationship in the other direction.

68
 Statistical Methods

Figure 6.2: One-tailed test:

Explanatory example:

We may wish to compare the mean of a sample to a given value „x‟ using a
t-test.

Our null hypothesis is that the mean is equal to x.

A two-tailed test will test both if the mean is significantly greater than x and
if the mean significantly less than x.

The mean is considered significantly different from x if the test statistic is in

the top 2.5% or bottom 2.5% of its probability distribution, resulting in a p-
value less than 0.05.

A one-tailed test will test either if the mean is significantly greater than x or
if the mean is significantly less than x, but not both.

69
 Statistical Methods

Then, depending on the chosen tail, the mean is significantly greater than or
less than x if the test statistic is in the top 5% of its probability distribution or
bottom 5% of its probability distribution, resulting in a p-value less than 0.05.

The one-tailed test provides more power to detect an effect in one direction
by not testing the effect in the other direction.

Hypothesis testing for a proportion:

Definition:

Hypothesis testing for a proportion is used to determine if a sampled

proportion is significantly different from a specified population proportion.

Explanatory example:

If you expect the proportion of male births to be 50 percent, but the actual
proportion of male births is 53 percent in a sample of 1000 births. Your aim
will be to test if this is significantly different from the hypothesized
population parameter.

Steps to perform a hypothesis testing for a proportion:

 Step 1:

Formulate your research question.

 Step 2:

Check to see if assumptions are met.

70
 Statistical Methods

 Step 3:

State the null hypothesis and the alternative hypothesis.

 Step 4:

Set an appropriate significance (alpha) level. By definition, the α-level is the

probability of rejecting the null hypothesis when the null hypothesis is true.

 Step 5:

Calculating the test statistic (z):

√ [ ]

P1 = Sample proportion.

P0 = Hypothesized population proportion.

n = sample size.

 Step 6:

Convert the test statistic to a p value.

 Step 7:

Decide between null and alternative hypotheses.

 Step 8:

State a conclusion about the research question.

71
 Statistical Methods

Explanatory example:

 Step 1

Research question:

Is the proportion of babies born male different from 50 percent?

 Step 2:

To test this claim, 1000 deliveries were surveyed using „simple random
sampling‟. In this sample, 530 gave birth to male boys.

 Step 3:

H0: p = 0.5

H1: p > 0.5

This is a one-sided (right-tailed) test because we want to know if the

proportion is greater than 0.5

 Step 4:

Alpha-level (α) is set at 0.05

72
 Statistical Methods

 Step 5:

√ [ ]

73
 Statistical Methods

 Step 6:

Convert the test statistic to a p value.

Figure 6.3:

Our z value = 1.894

According to the table (Appendix 2), P (z<1.894) = 0.9678

Therefore, P (z ≥ 1.894) = 1.0000 – 0.9678 = 0.0322

74
 Statistical Methods

 Step 7:

As p value < 0.050, our decision is to reject the null hypothesis.

 Step 8:

Conclusion about the research question:

There is a statistical evidence to state that the proportion of babies born

male is different from 50%.

75
 Statistical Methods

Practical exercises

1. A type I error occurs when we:

a. Reject H0, when H0 is actually true.
b. Do not reject H0, when H0 is actually true.
c. Reject H0, when H0 is actually false.
d. Do not reject H0, when H0 is actually false.

2. A type II error occurs when we:

a. Reject H0, when H0 is actually true.
b. Do not reject H0, when H0 is actually true.
c. Reject H0, when H0 is actually false.
d. Do not reject H0, when H0 is actually false.

3. The probability of making a type I error is denoted by:

a. Alpha.
b. Beta.
c. One minus alpha.
d. One minus beta.

4. The most frequently used significance (alpha) level is:

a. 0.100
b. 0.050
c. 0.010
d. 0.001

76
 Statistical Methods

5. The probability of making a type II error is denoted by:

a. Alpha.
b. Beta.
c. One minus alpha.
d. One minus beta.

6. The power of a test is:

a. Alpha.
b. Beta.
c. One minus alpha.
d. One minus beta.

7. You want to test if the mean serum bilirubin is higher in a group of viral
hepatitis patients as compared to a control group. This is:
a. One-tailed test.
b. Two-tailed test.

8. You want to test if the mean age is higher or lower in a group of viral
hepatitis patients as compared to a control group. This is:
a. One-tailed test.
b. Two-tailed test.

77
 Statistical Methods

Answers:

1. A.
2. D.
3. A.
4. B.
5. B.
6. D.
7. A.
8. B.

78
 Statistical Methods

Estimation statistics
Estimation statistics

“Estimation statistics” is a way of saying that you are estimating population

values based on your sample data.

This is performed by using inferential statistical methods such as the

“Confidence Interval”.

Confidence Interval (CI):

Definition:

A confidence interval consists of a range of values together with a percentage

that specifies how confident we are that the parameter lies in the interval.

A confidence interval is an indicator of your measurement's precision.

It is also an indicator of how stable your estimate is (a measure of how close

your measurement will be to the original estimate if you repeat your
experiment).

Formula:

CI = mean ± margin of error

79
 Statistical Methods

Margin of error:

Definition:

The maximum error of estimation (= the margin of error) is an indicator of the

precision of an estimate and is defined as one-half the width of a confidence
interval.

The 6-Steps to calculate CI:

 Step 1:

Write down the phenomenon you'd like to test.

 Step 2:

Select a sample from your chosen population.

 Step 3:

Calculate your sample mean and sample standard deviation.

 Step 4:

Choose your desired confidence level. The probability used to construct the
interval is called the level of confidence or confidence coefficient.

 Step 5:

Calculate your margin of error.

80
 Statistical Methods

 Step 6:

State your confidence interval.

Explanatory example:

 Step 1:

Say that the mean body weight of a male student in Mansoura University is 70
kg and the standard deviation is 10 kg (70 ± 10 kg). You'll be testing how
accurately you will be able to predict the weight of male students in Mansoura
University within a given confidence interval.

 Step 2:

You have randomly selected 400 male students.

 Step 3:

The sample mean was 70 kg and sample standard deviation was 10 kg.

 Step 4:

The most commonly used confidence levels are 90%, 95% and 99%.

Let's say you've chosen 95%.

81
 Statistical Methods

 Step 5:

Calculating the margin of error:

Margin of error = Critical value * Standard error of the mean (SEM)

= (Za/2) * σ/√(n)

Critical value = (Za/2) where a = confidence level

As confidence level chosen is 95% = 0.95

So, a/2 = 0.95 / 2 = 0.475

Check out the z-table (see appendix) to find the corresponding value that
goes with 0.475.

You will see that the closest value is 1.96, at the intersection of row 1.9 and
the column of .06.

82
 Statistical Methods

Figure 7.1: Z-table

83
 Statistical Methods

Standard error of the mean (SEM) = σ/√(n) where σ = Standard deviation and n
= sample size.

So, SEM = 10/√(400)

= 10/20

= 0.5

Therefore,

Margin of error = 1.96 * 0.5

= 0.98

 Step 6:

The 95% Confidence interval = 70 ± 0.98

So, the Confidence limits (lower and upper boundary values of the interval)
are

Lower bound = 70 – 0.98

= 69.02

Upper bound = 70 + 0.98

= 70.98

This means that we are 95% confident that the population mean lies in the
interval 69.02–70.98

84
 Statistical Methods

Practical exercises

1. The margin of error is calculated by multiplying the critical value by:

a. Mean.
b. Standard deviation.
c. Standard error of the mean (SEM).
d. Range.

2. If the confidence level chosen is 95%, the critical value is the value in z-
table that goes with:
a. 0.05
b. 0.95
c. 0.475
d. 0.095

3. Upper and lower boundaries of interval of confidence are classified as

a. error biased limits
b. marginal limits
c. estimate limits
d. confidence limits

4. If point estimate is 8 and margin of error is 5 then confidence interval

is:
a. 3 to 13
b. 4 to 14

85
 Statistical Methods

c. 5 to 15
d. 6 to 16
5. To develop interval estimate of any parameter of population, value
which is added or subtracted from point estimate is classified as
a. margin of efficiency
b. margin of consistency
c. margin of biasedness
d. margin of error
6. for a chosen confidence level of 95%, the critical value is:
a. 1.63
b. 1.74
c. 1.85
d. 1.96
7. In a study involving 100 participants, the mean height was 160 cm and
the standard deviation was 10 cm. Considering 95% confidence interval
(with a critical value of 1.96), the margin of error is:
a. 0.98
b. 1.96
c. 3.92
d. 5.88
8. In a study involving 100 participants, the mean height was 160 cm and
the standard deviation was 10 cm. Considering 95% confidence interval
(with a critical value of 1.96), the confidence limits are:
a. 159.02 and 160.98
b. 158.04 and 161.96
c. 156.08 and 163.92

86
 Statistical Methods

d. 154.12 and 165.88

Answers:

1. C.
2. C.
3. D.
4. A.
5. D.
6. D.
7. B.
8. B.

For questions 7 and 8:

SEM = 10 / 100

= 10 /10

=1

So, margin of error = Critical value * SEM

= 1.96 * 1

= 1.96

So, confidence limits = Mean ± Margin of error

= 160 ± 1.96

87
 Statistical Methods

= 160 – 1.96 and 160 + 1.96

= 158.04 and 161.96

Appendix A

88
 Statistical Methods

89
 Statistical Methods

90
 Statistical Methods

Appendix B

91
 Statistical Methods

92
 Statistical Methods

93
 Statistical Methods

94
 Statistical Methods

Appendix C

95
 Statistical Methods

96
 Statistical Methods

97
 Statistical Methods

98
 Statistical Methods

Appendix D

99
 Statistical Methods

Reference

1- Kaur P, Stoltzfus J, Yellapu V. (2018). Descriptive statistics. Int J Acad

Med 2018;4:60-3
2- Freud RJ, Wilson WJ and Mohr DL (2010). Statistical Methods (Third
Edition). AMSTERDAM, BOSTON, HEIDELBERG, LONDON NEW YORK,
OXFORD, PARIS, SAN DIEGO SAN FRANCISCO, SINGAPORE, SYDNEY,
TOKYO. Elsevier Inc.
3- Yan, F., Robert, M., & Li, Y. (2017). Statistical methods and common
problems in medical or biomedical science research. International
Journal of Physiology, Pathophysiology and Pharmacology, 9(5), 157–163.
4- Hollander, M. (2014). Nonparametric statistical methods (Third
edition.). Hoboken, New Jersey: John Wiley & Sons, Inc.
5- SPINA, D. (2011). Statistical methods in research. Methods Mol Biol.
2011; 746: 443-472.
6- Agresti, A. (1984). Analysis of ordinal categorical data. New York: Wiley.
7- Bancroft, T. A. (1968). Topics in intermediate statistical methods.
Ames, IA: Iowa State University Press.
8- Bancroft, T. A. (1968). Topics in intermediate statistical methods.
Ames, IA: Iowa State University Press.

100
 Statistical Methods

9- https://onlinecourses.science.psu.edu/stat500/sites/onlinecourses.scie
nce.psu.edu.stat500/files/lesson14/summary_table/index.pdf. Accessed
online on 07-10-2018.
10- Bhaskar, S. B., & Manjuladevi, M. (2016). Methodology for research
II. Indian Journal of Anaesthesia, 60(9), 646–651.
http://doi.org/10.4103/0019-5049.190620
11- Ali Z, Bhaskar SB. (2016). Basic statistical tools in research and data
analysis. Indian J Anaesth. 2016;60:662–9.

Book Coordinator ; Mostafa Fathallah

General Directorate of Technical Education for Health

https://www.facebook.com/mostafa.fathallah.3

101