CN 12

Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 2

ESTABLISHING VALIDITY AND RELIABILITY OF THE INSTRUMENT

II.OBJECTIVES:

●Discuss validity and reliability and ways of assessing validity and reliability of an instrument

●Summarize the concept of validity and reliability using diagram

●Write the importance of assessing validity and reliability of an instrument

●Validity refers to a degree to which the instrument measures what it intends to measure. It involves
collecting and analyzing data to assess the accuracy of an instrument. Here are ways to assess the
validity of a set of measurement:

●Face Validity. A subjective type of assessment. This is the simplest and easiest type of validity where
the validator skim the surface of the instrument in order to form an opinion. Moreover, it is often
criticized as the weakest form of validity (Stephanie, 2015).

●Content Validity. It refers to the appropriateness of the content of an instrument. A type of validity
that most often includes expert or people familiar with the construct being measured. The experts make
judgement about the degree to which the items in the questionnaire match the objective of the study.

●Criterion Validity. This type of validity measures how well the relationship between a measure and an
outcome. It can be measured in three ways:

●Convergent validity. Shows instruments are highly correlated with instruments measuring similar
variables. (e.g. geriatric suicide correlated to depression)

●Divergent validity. Shows the instrument is poorly correlated to instruments that measure different
variables. (e.g. correlation is low in instrument measuring motivation and instrument measuring self-
efficacy)

●Predictive validity. The instrument that has correlation with future criterion. (e.g. score of high efficacy
test related to performing task should predict the likelihood of completing task)

●Construct Validity. Defines how well a test measures what it claims to measure. It is used to know
whether the operational definition of a construct aligns to the true theoretical meaning of a concept.
There are three types of construct validity:

●Homogeneity. This means that the instrument measures one construct only.

●Convergence. The instrument measures construction similar to other instruments.


●Theory evidence. This is when behavior is similar to theoretical propositions of the construct measured
in the instrument

ESTABLISHING THE RELIABILITY OF THE QUESTIONNAIRE

Reliability refers to how accurate and precise the measuring instrument is. It yields consistent responses
over repeated measurements. In order to have a reliable instrument, you need to have a question that
yields consistent scores when asked repeatedly. Here are ways to assess the reliability of an instrument:

• Stability or Test-retest reliability. This is the simplest type of reliability where the same questionnaire is
administered twice and correlation between two set of score is computed.

• Split-half method. Also called equivalent or parallel forms. This is done by administering two different
set of questionnaires but with same topic and correlation between two set of score is computed.

• Internal consistency. This is when the instrument measures a specific concept. It is estimated based on
a single form of test administered on a single occasion.

Pilot testing of the questionnaire

Once you have done assessing the validity and reliability of the instrument, the next step to take is to
pilot test the questionnaire before distributing it to the target respondents of the study. Pilot testing is
like pre-testing the instrument. You may find 10-15 people to answer the questionnaire. In this process,
participants could put remarks on some questions. This could help you to enhance your questions.

Revise the questionnaire

After identifying some problem in your questionnaire, revise the questionnaire based on the feedback of
the participants during pilot testing. However, do not forget that the questionnaire should match the
research objective

You might also like