Lecture-Chapter 1

1. Statistics is defined as an art and science of (i) Collection of data, (ii) analyzing data and (iii) Interpreting data.
2. Data: Data consists of information coming from observations, counts, measurements, or responses.

Example: Age of the patients who come to the Hospital. Age is also called a variable as it changes from patient to patient.

How many variables are there in the hospital data? The data are information for each variable.

3. Population: A population is the collection of all outcomes, responses, measurements. For Hospital data, if we consider the ages of the patients, then all patient’s ages for all years of its services constitute the population of ages of patients. Similarly, the other data for all years of service will constitute populations.

4. Sample: A sample is a subset of population. For hospital data, the ages of the patients for one year (from October 2002 to September 2003) will constitute a sample.

5. Parameter: A Parameter is a characteristic of a Population. The average value of the age in the population usually denoted by µ (for mean) is a parameter.

6. Statistic: A Statistic is a characteristic of a sample. Such as the average age of the patients in a sample of patients from the population.

Data
Elements Variables Observations

Types of Data: Level of Measurements Qualitative Data: Quantitative data
Qualitative data consists of attributes, labels or non numerical entries
Quantitative data consists of numerical measurements or counts.

Example:
In the Hospital data: the variables are: Gender, Age, Stay overnight, number of visits, Amount of time, Amount of money, type of ailment.

Qualitative data are Gender, Stay overnight, type of ailment.
Quantitative data are Age, number of visits, Amount of time, amount of money.

Level of measurements:

Nominal level of measurements: Names, labels, qualities Gender, Ailments, Stay overnight.

Ordinal level of measurements: Data can be ordered Data could be qualitative or Quantitative Grades A, B, C, D can be ordered but the differences between the data entrée do not make sense.
Interval level of measurement These are quantitative Data can be ordered, can calculate meaningful differences between data entrees, Zero simply represents a position on the scale, does not mean emptiness, nothingness. Zero is not an inherent zero.
Ratio level of measurement: Similar to Interval level of measurements. Zero entry is an inherent zero. One Da ta value can be expressed as a multiple of the other data entrée.

EMERGENCY room patient’s data

OCT-02 - SEPT 2003

GENDER
AGE
STAY OVERNIGHT VISITS/YR
Hours
money
Ailment
1
F
47
NO
9
3
100
ALLERGY
2
F
32
NO
9
2
125
ALLERGY
3
F
18
NO
17
1
337
ALLERGY
4
M
20
NO
1
2
385
ALLERGY
5
F
22
NO
1
3
319
INFLAMMATION
6
M
41
NO
3
3
230
DENTAL
7
F
22
NO
4
2
98
NOSE BLEED
8
M
48
NO
3
1
125
NOSE BLEED
9
F
27
NO
2
2
386
BROKEN BONE
10
M
19
NO
1
3
448
BROKEN BONE
11
F
39
NO
2
2
715
BROKEN BONE
12
F
32
NO
6
3
719
BROKEN BONE
13
M
25
NO
1
1
449
INFLAMMATION
14
M
30
NO
1
1
125
BACK PROBLEM

Nominal data : Gender, Ailment type Ordinal data :Level of severity:
Fatal injury =1, severe injury=2, moderate injury=3,
Minor injury=4
Ranked data : All possible causes of death along with the number of death in USA:
Ten leading causes of death in USA in 1992
Rank
Causes of death
Total deaths
1
Diseases of the heart 717,706
2
Malignant neoplasm 520,578
3
Cerebrovascular disease 143,769
4
Chronic obstructive pulmonary disease 91,938
5
Accidents and adverse effects 86,777
6
Pneumonia and influenza 75,719
7
Diabetes mellitus 50,067
8
HIV infection