MEASUREMENT

MEASUREMNT
WHAT IS MEASUREMENT?
 It means assigning numbers or
symbols to objects or characteristics of
objects according to certain specified
rules.
WHAT IS MEASURED?
 Physical as well as abstract concepts.
MAPPING RULES
 Classification: Numbers are used to
group or sort responses. No order.
 Order: Numbers are ordered.
 Distance: Differences between
numbers are ordered.

 Origin: The number series has a
unique origin indicated by number

zero.
DATA AND SCALING
TYPES
 Nominal
 Ordinal
 Interval
 Ratio
NOMINAL DATA
 Nominal data are information on a variable
that naturally or by design can be grouped
into two or more categories that are
mutually exclusive and collectively
exhaustive.
 Nominal scale is a figurative labeling
scheme in which the numbers serve only as
labels or tags for identifying or classifying
objects.
PROPERTIES OF NOMINAL
SCALE
 Least powerful level of measurement.
 Indicates no order or distance relationship.
 No arithmetic origin.
 Used in surveys when data are being
classified by major sub-groups of
population.
 Cross-tabulation provides useful
information.
STATISTICS USED IN
NOMINAL SCALE
 Percentage, mode, chi-square,
binominal tests.
ORDINAL SCALE
 It is a ranking scale in which members
are assigned to objects to indicate the
relative extent to which the object
possess such characteristics.
 It indicate relative position not the
magnitude between the objects.

CHARACTERISTICS OF
ORDINAL SCALE
 It follows transitivity postulate: If a is greater
than b and b is greater than c, then a is
greater than c.
 Real difference between two ranks is not
known.
 Intervals of the scale may not be same.
 Addition or taking averages of the ranks are
technically incorrect.
STATISTICS USED IN
ORDINAL SCALE
 Appropriate measure of central
tendency: Median
 Dispersion: Percentile or quartile
measure
 Rank order correlation
 Measures of statistical significance:
Non parametric methods

INTERVAL SCALE
 Interval data have the power of nominal &
ordinal data plus the concept of equality of
interval.
 Location or zero point and unit of
measurement is arbitrary.
 It is not meaningful to take ratios of scale
values.
 Ratios of difference between the scale values
are permissible as they are independent of
the location of the zero point.
STATISTICS USED IN
INTERVAL SCALE
 Central tendency: Arithmetic Mean
 Dispersion: Standard Deviation
 Product moment correlation
 Statistical significance: t-test, F-test
and other parametric tests

 For skewed distribution: Median
(Central Tendency), Interquartile

ranges (Measures of Dispersion)
RATIO SCALE
 It possesses all the properties of
nominal, ordinal and interval scales,
and, in addition an absolute zero point.
 It allows to identify, classify objects,
rank order the objects, compare

intervals or differences, compute ratios
of scale values.
RATIO SCALE:
STATISTICS USED
 All statistics used in interval scale
 Specialised calculations- geometric
mean, harmonic mean, coefficient of

variation
SOURCES OF
MEASUREMENT
DIFFERENCES
 Respondent: Fatigue, Boredom, Ignorance,
Reluctance
 Situation: Presence or interference of other
person, noise
 Measurer: Rewording, paraphrasing,
conscious or unconscious prompting with
smiles, encouraging or discouraging etc.
 Data Collection Instrument: Too confusing
and ambiguous, poor selection of items
CHARACTERISTICS OF
SOUND INSTRUMENT
 Reliability
 Validity
 Practicality
VALIDITY
 It refers to the extent to which a test
measures what it actually wishes to
measure.
 Types of Validity:
1. Content,
2. Criterion-related,
3. Construct
CONTENT VALIDITY
 The content validity of a measuring
instrument is the extent to which it
provides adequate coverage of the
investigative question under study.
 Determination of Content validity:
Judgmental or panel evaluation
CRITERION-RELATED
VALIDITY
 It reflects the success of a measuring
instrument used for prediction or
estimation.
 Two types- Predictive and
Concurrent
 They differ only in a time perspective.
CRITERION-RELATED
VALIDITY
 Predictive: Prediction of the future;
criterion data are measured after the
passage of time.
 Concurrent: Description of the
present; criterion data are available
at the same time as predictor scores
 Method of determination: Correlation
CONSTRUCT VALIDITY
 Attempts to identify the underlying
construct(s) being measured and determine
how well the test represent it (them).
 Method of determination:
1. Correlation of proposed test with
established one
2. Convergent-discriminant analysis
3. Factor analysis
4. Multitrait-multimethod analysis
CONSTRUCT VALIDITY
1. Convergent Validity is the extent to which
the scale correlates positively with other
measures of the same construct.
2. Discriminant Validity is the extent to which
a measure does not correlate with other
constructs from which it supposed to differ.
3. Nomological Validity is the extent to which
the scale correlates in theoretically
predicted ways with measures of different
but related constructs.
RELIABILITY
 Reliability refers to the consistency of
scores obtained by the same person with
the same test on different occasions, or with
different sets of equivalent items, or under
other variable examining conditions.
 Reliability has to do with accuracy and
precision of a measurement procedure.
COEFFICIENT OF
RELIABILITY
 Stability
 Equivalence
 Internal Consistency
COEFFICIENT OF
RELIABILITY: STABILITY
 Test- Retest: Reliability of a test

instrument inferred from examinee
score. Same test is administered twice
to same subjects over an interval of
less than six months.
 Method of Determination: Correlation
COEFFICIENT OF
RELIABILITY:EQUIVALENCE
 Type: Parallel Forms
 Degree to which alternate forms of the
same measure produce same or

similar results. Administered
simultaneously or without a delay.
 Interrater estimates of the similarity of
judges’ observations or scores.

 Method of determination: Correlation
COEFFICIENT OF RELIABILITY:
INTERNAL CONSISTENCY
 Types: Split-half, KR20, Cronbach’s
Alpha
 Degree to which instrument items are
homogenous and reflect the same

underlying construct.
 Method of determination: Specialised
Correlational formulas
PRACTICALITY
 Economy: Trade-off occurs between
ideal research project and the budget.
 Convenience: Clear instructions, Easy
to administer with proper deign and

layout.
 Interpretability: Clear instruction about
how to interpret the test result.

INTERPRETABILITY
 Test manual should contain
1. Function of the test
2. Instruction for administration
3. Scoring keys and instructions
4. Norms for appropriate reference groups
5. Evidence about reliability
6. Guides to test use

MEASUREMENT

Uploaded by

Copyright:

Available Formats

MEASUREMENT

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

MEASUREMENT

Uploaded by

Copyright:

Available Formats

MEASUREMNT

 Distance: Differences between

numbers are ordered.

unique origin indicated by number

magnitude between the objects.

 Measures of statistical significance:

Non parametric methods

 Product moment correlation

 Statistical significance: t-test, F-test

and other parametric tests

(Central Tendency), Interquartile

rank order the objects, compare

mean, harmonic mean, coefficient of

 Test- Retest: Reliability of a test

same measure produce same or

judges’ observations or scores.

homogenous and reflect the same

to administer with proper deign and

how to interpret the test result.

You might also like