Latest posts

DEVELOPMENT OF MEASUREMENT TOOL

Research studies conducted in nursing measure every conceivable phenomenon and communicate findings to others in a concise and expedient manner.The methods of measurement commonly used are self-reports, observations and physiologic markers. Out of thousands of measurements tools and instruments, the task of the researcher is to choose or develop instruments that accurately, precisely, sensitively measure the variables of interest. The instrument of research should be carefully chosen. Though a large number of instruments are  available, it may be difficult to locate the exact instrument needed for a research study.

The development of a good instrument is an intensive task and should only be undertaken when an adequate, previously developed instrument is not found. The technique of developing measurement tool involves certain definite steps.


General outline in developing research instrument

The steps involved in developing a research instrument are:

Identify the concept to be measured: 
                                                     The most critical step in instrument development is identification and specification of the research concept. The researcher should be thoroughly familiar with all the literature on the concept. This allow the researcher to determine whether a suitable and adaptable instrument exists and ensure that all the dimensions of interest are tested. Once the researcher has determined the parameters of the concept, an outline should be developed. For examples, a researcher is interested in examining the concept of lies. Lying may be examining from many view points. An inclusive instrument would address each of these different approaches. The outline may be as follow:

  • Philosophical aspects of lying.
  • Ethical aspects of lying.
  • Moral aspects of lying.
  • Religious aspects of lying.
  • Professionals aspects of lying.
The concept development step is more apparent is theoretical studies than in pragmatic research, where fundamental concept are already established.

Determine the format of the measurement:
                                                               The researcher's knowledge of the concept and the respondent's knowledge of the concept will influence the choice of format. The researcher will also want to consider, how the information will be handled and what results are needed. For example, an unstructured interview may result in transcript of words that will have to be content analyzed. Though it may give a rich data st, it may be measured only at a nominal level. A questionnaire on the same topic may mean less rich data, but can give data measured at the internal/ratio level.
Develop the items: 
                         Using previously developed items, the researcher developed the items. The researcher must decide the format, for example, in case of questions- open ended or closed ended. Open ended questions do not give any specific response and may be long and difficult to categorize, closed ended questions result only in short answers and are easier to code and analyze. The number of questions too small number may not capture the subject, and too many may fire the respondent and may fail to answer the questions. The wording also should be considered the researcher who knows the concept very well, should be able to translate the knowledge into items that are understood by the respondent. The items should be worded concisely and simply. The researcher should consider reading level of the responding if a pen-paper instrument is developed. The wordings should be carefully phrased so that it will bias the respons by the sample.

Sequence the items:
                            Questions that ask about sensitive topics, such as income, sexual habits, should be placed at the end of the questionnaire. Respondents may not answer these questions. In such situations, the researcher can salvage some data. There should be logical sequence in the questionnaire -  simple to complex, less sensitive to more sensitive, most interesting to least interesting. Questions about different topics may be asked together. If there is no logical sequence random is best choice.  

Write the directions:
                          Respondents need carefully constructed, clear directions if they are to complete the instrument correctly. This is true for interview schedule. Questionnaires also require a cover letter if they are to be mailed. The covering letter should carefully state the purpose of study information about confidentiality, what the respondent should do with the questionnaire when it is completed and how the respondent may obtain further information.

Develop draft instrument and supporting materials:
                                                                           Once all the elements are ready, the instrument should be assembled. The 'look' of the instrument is also important when given to respondents. Such things as spacing the question, colour of the paper all can affect responses.

Review and pre-test:
                          The instrument should be reviewed by other knowledgeable researchers. The researcher should be asked to examine the instrument for completeness, clarity and layout. archrs. Content experts (both lay and professional) should be asked to comment on the instrument. After making needed changes, the questionnaire should be pre-tested using a small sample that possess characteristics similar to those proposed for  a larger study, under similar circumstances of proposed study. The pre-test helps the researcher to determine if the respondents can understand the items, directions are clear, and testing validity and reliability. It may be helpful to interview respondents after the pre-testing to elicit their suggestions and opinions.

Revise:
       Once pre-test is complete, the researcher should revise the instrument as indicated. If the researcher finds it necessary to make major revisions, a second pilot test should be performed.




No comments