Characterstics of Research Tool

                               CHARECTERISTICS OF RESEARCH TOOL

An ideal measuring instrument is one which results in measures that are relevant, accurate objective, sensitive and efficient. Measures which are physical or physiological have a higher chance of success in attaining these goals than measures that are psychological and behavioural. A number of techniques have been developed to evaluate the quality of the measuring tool to minimize error.

VALIDITY:

Refer to the degree to which the tool measures, what it is intended to measure. It refer to whether the instrument or scale quantifying what its claims to. E.g. weight scale measures the body weight; a tool which is valid for measure, need not be valid for another e.g. in the case of weight scale, it is not valid to measure temperature.

Major types of validity:

Different aspects or facets of validity questions are explained by the type of validity and number of indirect and logical methods have evolved to assess validity.

Content validity:

This is concerned with the sampling adequacy of the content area being measured. It refer to the fact that the items included in the tool represent a reasonable sampling of all possible items/ behaviours that make the domain being measured.

The content area to be tested must be studied to make certain that all major aspects or subcategories are represented in correct proportion. Experts in the content area are often called upon to analyse the items see if they adequately represent not only the content universe( or domain) but also in correct proportions. They also point out errors and ambiguities in items and make recommendation for changes.

Content validity assessment also involves a more elaborate procedure by which items are developed, given to experts for judgment and determination of percentage of items for each category. Percentage of agreement between the judges is consider as the basis for inclusion or rejections of items. After item analysis content validity needs to be assessed again to determine if the correct proportion of items for each subcategories has been maintained. If the proportion has not been maintain , exchange of items and sometime additions of items are made until the content is sampled appropriately while retaining an acceptable level of reliability.

Criterion related validity:

Seeks to establish that the measures correlates to another criterion rather than to ascertain how well the tool measures the particular phenomenon. E.g. the instrument measuring anxiety sate is said to be valid if the anxiety sores correlate with the same criterion, such as , person who are known to be anxious( e.g.. student prior to a difficult examinations) would be expected to have higher score on the anxiety scale. That is, groups known to possess or not to possess the characteristics being measured would be the criterion.

Sometimes distinctions are made between two types of criterion validity , predictive and concurrent.

Predictive validity ;

It refers to the adequacy of an instrument in predicting performance or behavioural of individuals on some future criteria. E.g. how well does the admission test predict the future performance of the students . this can be got by correlating admission test scores of first year nursing students with subsequent achievement in nursing in second, third , and fourth year. Here , the predictive validity of admission test of nursing is being evaluated.

Concurrent validity:  

on the  other hand , involve concurrent (i.e. simultaneous) measurement of the of the scale being validated. the difference between predictive and concurrent validity,then,is the difference n timing of obtaining measurements  on a criterian.one of the hardest aspects of criterion related  validity is selection and measurement of criterions. It must be relevant ,possess reliability and validity and free from biases

Construct validity;

In construct validity the investigator is concerned with the questions the concept under the investigator is adequately measured? Is there a valid bases for  inferring the course? This beings when the investigator formulate hypotheses  based on theory, about characteristic of those who have high scores on a test (compare to those with low sores) the measures can be dependent variable in some hypotheses and an independent variable in others evidence of construct validity is not established with the single study.

Variables like anxiety ,creativity, job satisfaction ,attitude towards numerous phenomena ,intelligence, empathy ,dependency, power, self concept etc are all constructs this variables have been constructed from theory and empirical data(i.e. qualitative data from research).for measures of variables that are constructs ,evidence must be presented that scores represent the degree to which an individual does indeed posses  or exhibit the construct. One common approach to construct validation is “known group technique” where the group which is expected to differ on the particular attribute because of some known  characteristics are administered the test. For e.g. in validating the tool to measure fear in pregnancy of primigravida, the researcher would also administer the tool to multigravidae.  Because one would logically infer that women who have not given birth ( primi) would express more fear than multigravidae.

Reliability:

 It refer to the accuracy and consistency of a measuring tool. A measure is reliable when an individual remains nearly the same in repeated measurements. Reliability is measured by reliability coefficient and the estimation of reliability can be done statistically.

How to estimate the degree of reliability?

Coefficient of stability shows consistency of performance on test over a period of time. It is based on the correlation between performance on initial test and retest after a distinct interval. That is, it involves the repetition of the same test.

A test may be reliable even though it may not be valid.

A valid test usually reliable.

Coefficient of equivalence shows consistency of performance on different form of a test. It is based on the correlation between performance on the different form of administered at nearly the same time. That is, a second equivalent form of the test is administered .

 Coefficient of internal consistency shows consistency of performance on different parts or items of the test taken at a single setting. This coefficient is usually computed by “ split half” correlation ( e.g. sperman- brown formula), or by using kunder- Richardson formula. That is, the analysis involves  subdivision of the test into two or more equivalent fractions.

Sensitivity and appropriateness:

Sensitivity refers to the capability to detect changes or differences when they to occur. Often , multimodal measurements are made because of the variations in the sensitivity  e.g. some fear reduction research show that changes in subjective feelings are not necessarily accompanied by changes in elevated heart rate. Heart rate, therefore, may not be sensitive enough to demonstrate change as a result of some interventions- particularly if it is observed immediately after administrations of the experimental variable. Therefore, it is necessary to include different to estimate fear.

Appropriateness refer to  whether or not the measure can be used for a particular target population. It also refers to the extent to which the target population can meet demands of the instrument, such as understanding and following the directions and having the physical stamina and motivation to complete the measure. If a measure is inappropriate if could measure wrongly.

Objectivity:

Objectivity means freedom from bias, a test should yields a clear scores value for each performance, the score being independent of the personal judgement of the score.

Economy:

Tests that can be given in short period of time are likely to gain the cooperation of the subject and to observer the time of those involved in test administration. The matter of expense is also a significant factor.

Relevance:

The degree to which the criteria established for selecting the item so that they conform to the aims of the measuring instrument.

Equality:

Achievement of the correct proportion among questions allotted to each of the teaching content.

Length:

The number of items in the test should depend of the content of the topic.

Practicability:

Simplicity of administration , scoring and interpretation are important factors in selecting test.

Interest:

Both the researcher and subjects should enjoy the experience of data collection. The investigator gains cooperation of subjects if responding to an instruments is interesting and enjoyable, and the measuring tool may not yield useful results if they are dull or seem silly.(foolish).

 .