VALIDITY OF MEASUREMENT S

1
VALIDITY OF MEASUREMENT S

• P M V Subbarao
• Professor
• Mechanical Engineering Department

Justification for Selection of Concepts to
Hardware ?????
2
How To Measure Any Other Property?

• It is essential to invent a scientific principle,
  which connects the property to be measured and
  physical displacement/length/ Just a Real Number.
• One needs to identify/ develop hardware which can
  work as per the scientific principle.
• What is the guarantee that the hardware exactly
  works as per the principle?
• How to develop high degree of confidence in a
  measurement?

3
Validity of Measurement

• When we decide to study a variable, we need to
  devise some way to measure it.
• Some variables are easy to measure and others are
  very difficult.
• We try to develop the best measures we can
  whenever we are doing research.
• A good measuring instrument or test is one that
  is reliable and valid.
• Let us look at test validity first.

4
Test Validity

• Test Validity refers to the degree to which a
  measuring strategy (instrument, machine, or test)
  measures what is to be measured.
• This sounds obvious right?
• A valid measure is the one that accurately
  measures the variable being studied.
• There are four/five ways to establish that your
  measure is valid
• Content validity
• Construct validity
• Predictive validity
• Concurrent validity
• Convergent validity and/or Discriminant validity.

5
Content Validity

• Content validity is established if your measuring
  instrument samples from the areas of skill or
  knowledge that compose the variable.
• This assumes that you have a good detailed
  description of the domain, something that's not
  always true.
• More the number of valid theories/skills, more
  will be the number of measurement strategies.
• Consider measurement of temperature
• Most popular valid theory for construction is ?

6
Coefficient of Thermal Expansion
material CTE (ppm/C)
silicon 3.2
alumina 67
copper 16.7
tin-lead solder 27
E-glass 54
S-glass 16
epoxy resins 15100
silicone resins 30300

• It is important to realise that
• The CTE is often not the same in all axes (that
  is, not isotropic).
• The CTE is rarely linear.
• The variation in CTE with temperature is only a
  fairly smooth function if the material is
  undergoing no phase transitions.

7

• Construct validity is the approximate truth of
  the conclusion that the measurement accurately
  reflects truth.
• The degree of translation of property to be
  measured into the measure.
• Construct validity is based on designing a
  measure that logically follows from a theory or
  hypothesis.
• Predictive validity, assesses the measurement's
  ability to predict something it should
  theoretically be able to predict.
• Concurrent validity, assesses the measurement's
  ability to distinguish between groups that it
  should theoretically be able to distinguish
  between.
• Refers to the ability of any measure to separate
  subjects who possess the attribute being studied
  from those who do not.
• Convergent validity assesses the degree to which
  the measurement is similar to (converges on)
  other measurements that it theoretically should
  be similar to.
• It is used when a valid measure exists for your
  variable but you want to design another measure
  that is perhaps easier to use or faster to take.
• Discriminant validity, examines the degree to
  which the measurement is not similar to (diverges
  from) other measurement that it theoretically
  should be not be similar to.

8
Ideas of Measurement Validation
9
Reliability

• Reliability is the consistency with which our
  measure measures.
• If you cannot get the same answer twice with your
  measure it is not reliable.
• A measuring strategy can be reliable and not
  valid, but if the instrument is not reliable it
  is also not valid.
• Measurement is never exact.
• At some point our measures always break down and
  errors creep into our data.
• This is when the concept of Error of Measurement
  becomes important.
• In order to be able to use any measure we need to
  know its error of measurement.

10
A good measuring strategy is reliable and,
because it is reliable, it has a small amount of
error in its observations.
The First Law of Measurements