Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
Manbhawan, Lalitpur
2019
CRITERIA OF GOOD MEASUREMENT/ GOODNESS OF
MEASURE
(RELIABILITY AND VALIDITY)
Reliability & validity as criteria of good measurement
validity and reliability are applied in both types of research: QUAN or
QUAL research
validity and reliability are addressed, using different instruments for
data collection.
Basic purpose of testing validity and reliability is to calculate
measurement error
Reliability is a necessary but not sufficient condition for validity in
research; reliability is a necessary precondition of validity, and
validity may be a sufficient but not necessary condition for reliability.
Four possible relations between reliability and validity:
i. Higher reliability but low validity
ii. Higher validity but low reliability
iii. Low validity and low reliability and
iv. High validity and high reliability
1. Validity
Validity as ascertaining the accuracy of the instruments of
data collection
Refers to truthfulness of findings
the extent that it measures what it purports to measure
In quantitative data validity might be improved through
careful sampling, appropriate instrumentation and
appropriate statistical treatments of the data.
In qualitative data validity might be addressed through the
honesty, depth, richness and scope of the data achieved,
the participants approached, the extent of triangulation
and the disinterestedness or objectivity of the researcher
Basic three types of validity: Content validity, Construct
validity and Criterion-related validity
Other are: face, internal-external validity etc.
1.1. Content Validity
concerns with the coverage of contents that meet the
objectives of a study
quality that covers the domain under the investigation
necessarily to be considered and careful sampling of
items is required while ensuring content validity
aims to "establish an instrument’s credibility, accuracy,
relevance, and breadth of knowledge regarding the
domain“
the experts' opinion or judgment are incorporated to
establish the content validity
subjective opinion from such experts establishes—or
does not establish—the content validity of the instrument
Content Validity Index Formula by Amin (2005) can also
be used to assess the content validity.
1.2. Construct Validity
an abstract; this separates it from the previous types of validity
which dealt in actualities – defined content
Seeks agreement between a theoretical concept and a specific
measures or procedures
the constructs that a researcher wants to measure by the measuring
tools
To establish construct validity the researcher would need to be
assured that his or her construction of a particular issue agreed with
other constructions of the same underlying issue
construct validity is addressed by convergent and discriminant
techniques
Convergent techniques imply that different methods for researching
the same construct should give a relatively high inter-correlation
Discriminant techniques suggest that using similar methods for
researching different constructs should yield relatively low inter-
correlations
The accumulated evidence collected from face, content, predictive,
concurrent, convergent and divergent validity also verify the
construct validity
1.3. Criterion-related Validity
Demonstrate the accuracy of a measure by comparing it with
another measure (eg. Questionnaire, observation, interview and
documentation)
Differentiates individuals on a criteria it is expected to predict
Two methods: Concurrent and predictive validity
Concurrent validity compares the results yielded from two
different instruments measuring the same constructs at the
same time- a test to meet the criteria at the same time
(correlational method or regression method)
Predictive validity describes the relation between the results of
first round test with the second round. It is achieved if the data
acquired at the first round of research correlates highly with
data acquired at a future date
Concurrent validity is very similar to its partner – predictive
validity – in its core concept (i.e. agreement with a second
measure); what differentiates concurrent and predictive validity
is the absence of a time element in the former; concurrence can
be demonstrated simultaneously with another instrument.
2. Reliability
Ensures the consistency of a measure to be yielded by an
instrument over a period of time in the same group or
over different groups at the same or different time.
Concerned with precision and accuracy
A synonym for dependability, consistency and replicability
Degree of consistency or dependability on the
measurement of a construct
Name as credibility, neutrality, conformability,
dependability, consistency, applicability, trustworthiness
and transferability in qualitative research
Three principal types of reliability: stability, equivalence
and internal consistency
Coefficient of correlation is 0.7 or more than this for the
(
acceptance of reliability Nunnally & Bernstein, 1994).
2.1. Reliability as stability
a measure of consistency over time and over similar
samples
a test and then a retest were undertaken within an
appropriate time span, then similar results would be
obtained.
The researcher has to decide what an appropriate length of
time is
Reliability as stability can also be stability over a similar
sample
a test or a questionnaire simultaneously to two groups of
students who were very closely matched on significant
characteristics
correlation coefficient by test/retest method can be
calculated either for the whole test by using the Pearson
statistic, Spearman or a t-test)
2.2. Reliability as Equivalence
achieved first through using equivalent forms (also
known as alternative forms) of a test or data-gathering
instrument.
When an equivalent form of the test or instrument is
devised and yields similar results, then the instrument
can be said to demonstrate this form of reliability.
Reliability can be measured through a t-test, through
the demonstration of a high correlation coefficient
Reliability as equivalence may be achieved through
inter-rater reliability
All researchers must be achieved, through ensuring
that each researcher enters data in the same way
Number of actual agreements
Inter-rater agreement as a percentage: 100%
Number of possible agreements
Reliability as internal consistency