Transcript Research and Evidence

CHAPTER 1
Introduction to statistics
What is Statistics?
•Statistics is the term for a collection of
mathematical methods of organizing,
summarizing, analyzing, and interpreting
information gathered in a study
Data and Data Analysis
We have two types of research study
•In quantitative research, data are usually
quantitative (numbers) and subjected to
statistical analysis. Mainly the data is
collected by close ended questions
•Qualitative research, data are usually
narrative and collected by open ended
questions
Example of close ended question (Likert scale) to measure
attitude toward mental illness
SA = Strongly agree
A = Agree
D = Disagree
SD = Strongly disagree
?? = Uncertain
Dr. Yousef Aljeesh
Items
Reflect the topic of the study
Strongly
agree
(5)
Agree
(4)
People who have had
Mental illness can
become normal and
productive citizens after
treatment .
Mental ill patient’s who have
been in Psychiatric hospital
or center should not be
allowed to have children.
Dr. Yousef Aljeesh
Uncertain
(diversity)
(3)
Disagree
(2)
Strongly
disagree
(1)
Example of open ended question
• What is the perception of you organization
towered female holding high managerial
positions?
………………………………………………………………………
………………………………………………………………………
………………………………………………………………………
………………………………………………………………………
………………………………………………………………………
Where Do Data Come From?
•Example 1: Interviews/questionnaires
–Question: On a scale from 0 to 10, please rate
your level of fatigue
–Answer (Data):




Person 1: 7
Person 2: 3
Person 3: 10
Etc.
Variables
A variable is something that takes on different
values
Example of variables –Height, sex, weight, age,
level of education, marital status, respiratory
rate, heart rate and etc…
Types of Variables
–Independent variable: The hypothesized cause
of, or influence on an outcome
–Dependent variable: The outcome of interest,
hypothesized to depend on, or be caused by
the independent variable
Research Questions
•Research questions communicate the research
variables and the population(the entire group
of interest)
–Example: In hospitalized children (population)
does music (IV) reduce stress (DV)?
Types of Sampling
1. probability Sampling
2. Non- probability Sampling.
Dr. Yousef Aljeesh
Probability sample
The probability sample means, the probability
of each subject to be included in the study.
There are four types of probability sample
Dr. Yousef Aljeesh
Four basic kinds of probability samples.
a. Simple random sample. The simple random sample
is the simplest probability sample, so that every element in
the population has an equal probability of being included.
Note
All types of random samples tend to be representative.
Dr. Yousef Aljeesh
b. Stratified random samples
In a stratified random sample, the population is first divided
into two or more homogenous strata (age, gender, occupation,
level of education, income and so forth) from which random
samples are then drawn. This stratification results in greater
representativeness.
Dr. Yousef Aljeesh
C. Cluster samples
For many populations, it is simply impossible
to obtain a listing of all the elements, so the
most common procedure for a large surveys
is cluster sampling.
Dr. Yousef Aljeesh
D. Systematic samples
Systematic sampling involves the selection of every (kth) element
from some list or group, such as every 10th subject on a patient
list. If the researcher has a list, or sampling frame, the following
procedure can be adopted. The desired sample size is started
at some number (n). The size of the population must be known or
estimated (N). By dividing (N) by (n), the sampling interval is the
standard distance between the elements chosen for the sample.
Dr. Yousef Aljeesh
Example
if we were seeking a sample of 200 from a population of 40,000,
then our sampling interval would be as follows:
K=
40,000 = 200
200
In other words, every 200 the element on the list would be sampled.
The first element should be selected randomly, using a table of
random numbers, let us say that we randomly selected number
73 from a table. The people corresponding to numbers 73, 273,
473, 673, and so forth would be included in the sample.
Dr. Yousef Aljeesh
2. Non-probability Sample
Non-probability sample is less likely than probability
sampling to produce a representative samples. Despite
this fact, most research samples in most disciplines
including nursing are non-probability samples.
Dr. Yousef Aljeesh
a. convenience sampling (Accidental, volunteer)
The use of the most conveniently available people or subjects in a study. For
example, stopping people at a street corner to conduct an interview is
sampling by convenience. Sometimes a researcher seeking individuals with
certain characteristics will stand in the clinic, hospital or community center to
select his convenience sample. Sometimes a researcher seeking individuals
with certain characteristics will place an advertisement in a newspaper, so the
people or subjects are volunteer to take
apart of the study.
Dr. Yousef Aljeesh
b. Snowball or network sampling
Early sample members are asked to identify and refer other
people who meet the eligibility criteria. or it begins with a few
eligible subjects and then continues on the basic of subjects
referral until the desired sample size has been obtained. This
method of sampling is most likely to be used when the researcher
population consists of people with specific traits who might
otherwise be difficult to identify.
Dr. Yousef Aljeesh
C. Quota Sampling
Quota sampling is another form of non-probability sampling.
The quota sample is one in which the researcher identifies
strata of the population and determines the proportions of
element needed from the various segments of the population,
but without using a random selection of subjects.
Dr. Yousef Aljeesh
Note:
Although there are no simple formulas that indicate how large
sample is needed in a given study, we can offer a simple piece of
advice: you generally should use the largest sample possible.
The larger the sample the more representative of the population it
is likely to be.
Dr. Yousef Aljeesh
Variable and constant
Variable: is something that varies or takes in different
values (weight, sex, blood pressure, and heart rate)
are all examples of characteristics that vary from one
person to the next. If they did not vary, they would
be constants
Discrete Versus Continuous
Variables
•Variables have different qualities with regard to
measurement potential
–Discrete variables
–Continuous variables
Note:
- We use non-parametric tests in case of
Nominal and Ordinal measurement
(Example: Chi-Square test)
- Both depend on percentages because Mean
does not make sense
Note
In interval scale, there is no real or rational
zero point
Another Example
Weight (Zero weight is actual possibility)
It is acceptable to say that some one who
weights 100 kg is twice as heavy as some one
who weights 50 kg.
Note
Interval and Ratio measurements are continuous
variables and parametric tests should be used in
this situation. Also Mean is applicable
Types of Statistical Analysis
• Calculation
–Manual versus computerized
• Purpose
–Descriptive versus inferential
• Complexity
–Univariate, bivariate, multivariate
Descriptive Statistics
•Researchers collect their data from a sample of
study participants—a subset of the population
of interest
•Descriptive statistics describe and summarize
data about the sample
–Examples: Percent female in the sample,
level of education, Income, residency and ect
Example 1 of Descriptive statistics
Distribution of study population according to place of work
Hospital name
Target
population
Respondents
Percentage
Response rate
Al-shifa hospital
56
51
35.7
91.07%
Nasser medical complex
21
21
14.7
100%
European Gaza hospital
21
17
11.9
80.95%
Aqsa Martyrs Hospital
14
14
9.8
100%
Kamal Adwan hospital
9
9
6.3
100%
Abu Yousef Al Najjar
12
8
5.6
66.6%
Beit Hanoun hospital
10
10
7.0
90.9%
Ophthalmic hospital
7
6
4.2
85.7%
Crescent Alemaraty
9
7
4.9
77.7%
159
143
100.0
Total
Calculation of Response Rate
Response Rate (RR) = Respondents
(R)
Target Population (TP)
RR=
51
56
100 = 91.07
100
Example 2 of Descriptive statistics
Distribution of Study Population According to Height, Weight and BMI
Variables
Height (cm)
Weight (kg)
Body Mass Index
(BMI)
Category
(N= 143)
Frequency
Percentage (%)
166cm and less than
41
28.7
167 – 176 cm
56
39.2
177 – 186 cm
40
28.0
187cm and above
6
4.2
Total
143
100.0
67kg and less than
32
22.4
68-78 kg
39
27.3
79-89 kg
41
28.7
90 kg and above
31
21.7
Total
143
100.0
Less than 25
55
0.7
22.5-29.5
33
37.8
30 and more
25
44.1
143
100.0
Total
Example 3 of Descriptive statistics
Age distribution
45.9
50
45
40
35
28.4
25.7
30
25
‫شرق‬
20
15
10
5
0
30 Yrs and less
From 31 to 45 Yrs
More than 45 Yrs
Example 4 of Descriptive statistics
Gender distribution
Example 5 of Descriptive statistics
Distribution of subjects by governorates
Items
No.
%
North
15
13.33
Khanyounis
13
11.6
Gaza
56
50
Rafah
13
11.6
Mid Zone
15
13.33
Total
112
100
Inferential Statistics
• Researchers obtain data from a sample but
often want to draw conclusions about a
population
• Inferential statistics are often used to test
hypotheses(predictions) about
relationships between variables
Example:- Positive, negative, directional hypothesis
and etc.
Example of inferential statistics
Association between socio-demographic factors and diarrhea
among children aged less than 5 years (N=140)
Diarrhea
Factor
Father age
OVC status
Type of family
Cases
N (%)
Control
N (%)
(20 – 30) years
33 (47.1)
37 (52.9)
(31 – 40) years
34 (48.6)
22 (31.4)
(41 – 59) years
3 (4.3)
11 (15.7)
Orphaned
1 (1.4)
2 (2.9)
Vulnerable
4 (5.7)
3 (4.3)
Not OVC
65 (92.9)
65 (92.9)
Nuclear family
46 (65.7)
50 (71.4)
Extended family
24 (34.3)
20 (28.6)
χ2
p value
7.371
0.025*
0.476
0.788
0.530
0.466
Hypotheses
Definition of hypothesis : It is a statement of
predicted relationship between two or more than
two variables.
Dr. Yousef Aljeesh
Types of Hypotheses
1.
Simple Hypothesis : A hypothesis that predicts the
relationship between one dependent variable (DV) and
one independent variable (IDV). It is easy to test and
analyze it.
Example
There is a relationship between smoking and development
of stroke among hypertensive patients in Gaza strip.
Dr. Yousef Aljeesh
2. Complex hypothesis: (Multivariate hypothesis) :
A hypothesis that predicts the relationship between two or
more dependent variables and two or more independent
variables.
Example:
There is a relationship between high fat diet and smoking
and development of atherosclerosis and stroke among
hypertensive patients in Gaza strip.
Dr. Yousef Aljeesh
3. Directional hypothesis: is one that specifies the
expected direction of the relationship between
variables. The researcher predicts not only the
existence of a relationship but also the nature of
the relationship.
Dr. Yousef Aljeesh
Example
1. There is a positive relationship between Smoking
and lung cancer
Dr. Yousef Aljeesh
4. Statistical hypothesis (Null hypothesis): is one
that stated there is no relationship between
variables.
Example
1. There is no relationship between Smoking and lung cancer
2. There is no relationship between obesity and Breast cancer.
Dr. Yousef Aljeesh