The Big Question. Questionnaire Reliability

A questionnaire is an important method of data collection used extensively in research. It is a fairly inexpensive tool and offers anonymity, thus making it a favorite tool of many researchers. While preparing a questionnaire, it is important to that it needs to be easy to understand, have an aesthetic layout, have precise framing of sensitive questions and be interactive for it to be an effective tool.

In order to get accurate results with a questionnaire, it needs to have two factors. Validity and Reliability.

Validity:
All the data collected from the academic questionnaire must be absolutely valid i.e. it must provide effective answers to the research question it was conducted for. It is basically the ability of the instrument to measure exactly what it was intended to measure.
If a questionnaire can’t seek answers for the research question, it becomes ineffective tool.

Reliability:
A questionnaire has to reliable i.e. it has to have the ability to produce reproducible results. In other words, a questionnaire needs to have the ability to consistently produce similar/same results repeatedly.
Each time a questionnaire is used, similar score should be obtained. Only then is a questionnaire considered reliable for academic research. It can’t be quantitatively measured, but can be estimated using correlation coefficients.

Reliability of Questionnaires is measured in the following aspects:

Stability

It basically ensures that the same results are obtained consecutively. It is generally tested using the Test-Retest method.
In the Test-Retest method, the same questionnaire is given to two or more people, one after the other to determine if they all get similar responses.

Internal Consistency

It ensures that all parts and sub-parts of the survey are used to measure the same characteristic. A survey needs to be homogenous to have internal consistency. It can be tested using the Split-half method.
Using the split-half method, the performance of a group of people is correlated on two equivalent halves of the same survey.

Equivalence

When a single phenomenon is studied simultaneously by two or more observers, they need to have equivalent conclusions. A survey is said to have equivalence when all who have simultaneously observed the test come to a similar conclusion. The Inter-Rater test is used to measure equivalence.
In this test, the percentage of agreement between multiple researchers is determined while they observe the same questionnaire.

Basically, a questionnaire has to be unbiased and have no ambiguities. It needs to have good quality of data that is accurate and credible. Reliability is a huge factor without which the finding of a questionnaire hold no credibility and serve no academic purpose.

When preparing a questionnaire, one must remember that a questionnaire can be reliable but not valid, but every valid questionnaire is generally reliable.

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s