Reducing the Items in the Research Instrument

There are instances that you need to reduce the items in your research instrument. If the items in the instrument that you have prepared are too many, chances are, your respondents would not read it thoroughly and eventually give you a poor result. Let’s admit that nowadays, respondents don’t like to answer too many questions. For them, it is a waste of time. With this in mind, how can you lessen the number of items in your questionnaire? Here are your options:

1. Validate the contents of your research instrument.

Aside from measuring the validity of the instrument, content validity can help you decide which items must be deleted, thus reducing the items in your questionnaire. Content validity can be done by the experts in that field that you are trying to investigate. But who are considered “experts?” These are the people who have doctorate degrees in that field and have practiced their profession for many years. Usually, a minimum of three persons must take a look at your instruments.

master student

However, if your paper is towards fulfilling a master’s degree requirement and there is no one who has a PhD in that area, those who have master’s degree can be taken as experts. If master’s degree holders are still unavailable, the panel of examiners may allow you to avail the help of those who have been practicing their profession for not less than ten years as “experts” in that field.

These people are knowledgeable as to the depth of the contents of the subject area that you want to investigate in and must be adept as to which items should be included to determine a particular variable (construct) and its sub-variable (sub-construct) in your study. They can also tell you which of the items from your questionnaire could be deleted.

2. Do factor analysis for data reduction.

If you choose to do a factor analysis, this will help you determine the construct validity of your instrument and help you decide, with the help of your statistician, which items are to be culled out from your instrument. To know more on this, see What are the Psychometric Properties of a Research Instrument?

3. Have both content and construct validity.

Having the two types of validity, I believe, will make your instrument better. As a researcher, you are the one to decide or with the help of your research adviser, which items must be deleted using the results of content and construct validity.

Generally, instruments which underwent the content and face validity are much shorter and can give more accurate results than one which did not. Thus, this gives you more confidence in the interpretation of your data.