Scribd
Upload
145 views6 pages

Nature of Veterinary Data Scale of Measurment, Data Elements. - R-019-1

1. The document discusses various types of veterinary data that epidemiologists investigate, including qualitative and quantitative data. Qualitative data describes group membership while quantitative data relates to amounts. 2. Several scales of measurement for data are described, including nominal, ordinal, interval, and ratio scales. Nominal data involves classification, ordinal includes ranking, interval includes equivalence and ratios of intervals, and ratio includes a true zero point. 3. Composite measurement scales are also discussed, which numerically combine different types of data, such as behaviors over time to assess pain in animals. Accuracy, refinement, precision, reliability, and validity are important aspects for qualitative and quantitative data and diagnostic techniques.

Uploaded by

Hero
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
145 views6 pages

Nature of Veterinary Data Scale of Measurment, Data Elements. - R-019-1

1. The document discusses various types of veterinary data that epidemiologists investigate, including qualitative and quantitative data. Qualitative data describes group membership while quantitative data relates to amounts. 2. Several scales of measurement for data are described, including nominal, ordinal, interval, and ratio scales. Nominal data involves classification, ordinal includes ranking, interval includes equivalence and ratios of intervals, and ratio includes a true zero point. 3. Composite measurement scales are also discussed, which numerically combine different types of data, such as behaviors over time to assess pain in animals. Accuracy, refinement, precision, reliability, and validity are important aspects for qualitative and quantitative data and diagnostic techniques.

Uploaded by

Hero
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 6

ASSIGNMENT

Submitted to: Dr. Sajid Hameed


Class: DVM-8
Session: 2017-2022
Submitted by: Manahil Arshad (19)
Department: FCV&AS
Subject: Epidemiology and public
health-02802
Nature of veterinary data scale of measurement, data elements
The epidemiologist investigates the frequency and distribution of disease (and sometimes other
characteristics, such as performance) in groups of animals.

Animals are put into the diseased category because they possess certain attributes, such as clinical
signs and lesions, that are used to define, and sometimes to name, the disease.

 Some data are observations; for example, the recording of diarrhea.


 Other data are interpretations of observations; for instance, a diagnosis, which represents an
interpretation of a set of clinical signs, lesions and laboratory results.

Classification of data

Additionally, some data that epidemiologists use nowadays are stored in computers as codes.

Data can be broadly classified into qualitative and quantitative

1. Qualitative data describe a property of an animal; that is, its membership of a group or class.
Such data therefore are termed categorical.
Examples are the breed and sex of an animal.
2. Quantitative data relate to amounts, rather than just indicating classes.
Examples are prevalence, incidence, body weight, milk yield, temperature and antibody titer.
These data may be further divided into discrete and continuous.
 Discrete data can have only one of a specified set of values, such as whole numbers (1, 2, 7, 9,
etc.)
Example the number of teats on a sow.
Discrete data generate counts. Thus, aggregates of qualitative (categorical) data are counted (e.g.
the total number of male dogs or Friesian cows).
 Continuous data may have any value within a defined range (though the range can be infinite).
Examples are the girth of a cow and its body weight.
Continuous data are usually quantified by comparison with a fixed unit; that is, they are
measured. Continuous data therefore generate measurements.

Scales (levels) of measurement


Although measurement can be applied strictly only to quantitative, continuous data,
'scale of measurement' is also conventionally used to describe the 'strength' of values that
can be attached to both qualitative and quantitative data.
 There are four main scales (levels) of measurement i.e.
1. Nominal
2. Ordinal
3. interval and ratio
4.visual analogue scale is also sometimes used.

1. The nominal (classificatory) scale


This scale involves the use of numbers (or other symbols) to classify objects.
Thus, male and female can be coded 1 and 2, respectively, and a non-veterinary example is the
use of numbers or letters on aircraft or car number plates to indicate their origin.
The property of a nominal scale is equivalence means members in a class must possess the same
property. The only legitimate operation that can be performed on a nominal scale is
transformation of the symbols.
For example, if the disease, diabetes, was numerically coded as 671, then this value could be
transformed or changed to 932 for all diabetics.
This scale is a 'weak' form of 'measurement'.

2. The ordinal (ranking) scale


The ordinal scale allows groups to be related to other groups.
Most commonly, the relation can be expressed in terms of equal to, greater than (») or less than
(«).
Examples are the use of body condition scores for sheep, cattle, horses, dogs and cats
and clinical grading of severity of disease, such as lameness.
The difference between the nominal and the ordinal scale is therefore that the ordinal includes
not only equivalence but also the 'greater than' and 'less than' property.
In the ordinal scale, any transformation must preserve the order. It does not matter what number is
attached to a class, as long as the relationship with other classes is consistent. Thus, a carcass
condition score scale can include 5 = good and 1 = poor and equally 1 = good and 5 = poor, as
long as the numbers between 1 and 5 maintain the same order of ranking.
Although 'stronger' than the nominal scale, the ordinal scale is still a relatively 'weak' form of
'measurement'.

3. The interval scale


In an interval scale, the distance between the ranked values is known with some accuracy.
Example is body temperature. Two thermal interval scales commonly are used - Celsius and
Fahrenheit - each containing the same amount of information.
The interval scale therefore includes equivalence, 'greater than' relationships, and ratios of
intervals. The interval scale is a relatively 'strong' form of (actual) measurement.

4. The ratio scale


The ratio scale is an interval scale with a true zero point.
Example Weight is a ratio scale.
The various weight ratio scales may have units of kilograms, grams, pounds or ounces, but they
all start from the same zero point. But a ratio scale is not necessarily associated with ratios, many
of which are ratios of counts i.e. Weight, for instance, is a ratio scale that is not a ratio.

The visual analogue scale


The visual analogue scale (VAS) uses a straight line, usually 10 cm long, the extreme limits of
which are marked with perpendicular lines. Both ends have a verbal description of each extreme
of the variable that is being measured.
Example of a VAS for recording severity of lameness in sheep. The VAS is commonly used in
human medicine to assess pain, and has been similarly applied in veterinary medicine.
Nature of ordinal and visual analogue scales
Ordinal and visual analogue scales are relatively simple, non-invasive ways of recording severity
of lesions and clinical signs, and so are attractive options for recording changes in patients' status
in clinical trials. However, both scales are subjective.
 These scales of measurement have been considered in detail because of their importance in
determining appropriate statistical methods. Most statistical tests can be used with interval and
ratio scales. However, not all tests can be applied to nominal, ordinal and visual analogue scales.

Composite measurement scales

The measurements described so far either comprise objective, single values (e.g., weight) or
single subjective values based on several qualitative criteria Scales may also be devised by
numerically combining nominal, ordinal, and interval and ratio data; these are termed composite
measurement scales (CMSs). for example, summarizes the components of a CMS for acute pain
in lambs following castration and tail-docking.
The scale - termed REQ scoring (named from the initial letters of its first and last components) -
sums the frequency of observation of various behavioral characteristics
over a period of time (usually 30-60 minutes) after the procedures were undertaken.

Data elements
Accuracy refinement, precision, reliability and validity These terms can be used in relation to
qualitative data (e.g., the description of a disease) and to quantitative measures (e.g. of prevalence
and weight).
Accuracy
Accuracy is an indication of the extent to which an investigation or measurement confirms to the
truth.
Refinement
The degree of detail in a datum is its refinement. Thus, 13 kg and 13.781 kg may both represent
the accurate weight of an animal, but the second record is more refined than the first. Increasing
the refinement of descriptive diagnostic data may improve their epidemiological value.
Precision
Precision can be used in two senses. First, it can be used as a synonym for refinement. Secondly,
it can be used statistically to indicate the consistency of a series of measurements.
Reliability
A diagnostic technique is reliable if it produces similar results when it is repeated. Thus
repeatability is a characteristic of a reliable technique.
Validity
If a diagnostic technique measures what it purports to measure, it is valid. Validity is a long-term
characteristic of a technique, of which sensitivity and specificity are indicators. The validity of a
technique depends upon the disease that is being investigated and the method of diagnosis
Representation of data: coding
Data are usually represented by both words and numbers. An alternative type of notation involves
coding. This is a means of representing text and numerals in a standardized, usually abbreviated,
form. Thus, it is easier and quicker to record the code number 274 than to write its possible
textual equivalent, 'contagious bovine pleuropneumonia'.
Following are the coding systems used to interpret data.
1. Numeric codes
2. Consecutive and hierarchic codes
3. Alpha codes
4. Alphanumeric codes

SNOMED

The most widely used multiaxial international coding system is SNOMED (Systematized
Nomenclature of Human and Veterinary Medicine) which was introduced as an exclusively
medical system, SNOP (Systematized Nomenclature of Pathology). An early version was
modified for veterinary use and the Systematized Nomenclature of Veterinary Medicine
(SNOVET), which used six axes: topography, morphology, an etiology, function, disease and
procedure.

You might also like