Statistical Methods Adopted in Psychometric Property of Instruments in Nursing Research

 

Dr. Sampoornam. W

Associate Professor, Mental Health Nursing Department, Dhanvantri College of Nursing, Pallakkapalayam, Namakkal, Tamilnadu, India

 

ABSTRACT:

In the research study platform, instrument is the key core aspect which directs the investigator to undertake reliability and validity. While developing research instrument, psychometric property in an active part where the scientist pay attention towards reliability and validity.  In other words, psychometric property refers to the reliability and validity of an instrument. Psychometrics is the construction and validation of measurement instruments and assessing if these instruments are reliable and valid forms of measurement. In behavioral medicine, psychometrics is usually concerned with measuring individual’s knowledge, ability, personality and types of behaviors.

 

 

INTRODUCTION:

Measurement usually takes place in the form of a questionnaire and questionnaires must be evaluated extensively before being able to state that they have excellent psychometric properties, meaning a scale is both reliable and valid (Grimm, G. L., and Yarnold, P. R. 2000). Hitherto psychometric properties are quantifiable attributes (e.g., validity, reliability) that relate to the statistical strength or weakness of a test or measurement. The psychometric property of a measure refers to the theoretical principles and rules as applied to measurement (Nunnally and Bernstein 1994).

 

Statistical tests used in Reliability and Validity-An Quantitative approach:

Reliability refers to the consistency while validity refers to the test’s accuracy.

 

An instrument should accurately measure what it ought to measure. In order to prevent the rigor and error of an instrument, reliability and validity should be carried out. The Hallmark of science is the pursuit of phenomena and the limitation of error.

 

Reliability:

Essentially, any research tool should provide the same information if used by different people (inter-rater reliability) or if it is used at different times, calculated by adopting Multirater Kappa. Test-Retest reliability is used to assess the consistency of a measure from one time to another, calculated using Pearson Product Moment Correlation. In Parallel Form reliability the consistency of the results of two tests constructed in the same way from the same content domain is analyzed by adopting Pearson Product Moment correlation coefficient and Spearman Brown if test length has been changed.

 

Internal consistency of items such as individual questions in a questionnaire can be measured using statistical procedures such as Cronbach’s alpha coefficient (Cronbach 1951), randomly splitting all the responses into two sets, totaling the scores on the two sets and working out the correlation between the two sets. This is known as ‘split-half’ test.

 

Validity:

Construct validity:

Construct validity refers to the degree to which inferences can legitimately be made from the operationalization in the study to the theoretical constructs on which those operationalizations were based. In general, Exploratory Factor Analysis (EFA); Confirmatory Factor Analysis (CFA) and Principal Components Analysis are used for the psychometric property soundness. 

 

The two broad types of Construct validity are

Translation validity:

Focus on whether the operationalization is a good reflection of the construct. This approach is definitional in nature, it assumes to have a good detailed definition of the construct and that can check the operationalization against it.

 

Face validity:

The instrument on the face of it appears to measure the construct under study.

 

Content validity:

Extent to which items in the tool sample the complete range of the attribute under study. Content Validity Index is an appropriate statistical method to test it.  

 

Criterion-related validity:

By using Pearson Product Moment Correlation, relationship linking the attributes in a tool with the performance on a criterion can be estimated.

 

Predictive validity:

The degree to which test scores predict performance on some future criterion(High Pearson Product Moment Correlations).

 

Concurrent validity:

Scores on the measurement tool are correlated to a related criterion at the same time (High Pearson Product Moment Correlations).

 

Convergent validity:

Extent to which different measures of the same construct correlate with one other (High Pearson Product Moment Correlations).

 

Discriminant validity:

Extent to which measures of different constructs correlate   with one other (Low Pearson Product Moment Correlations).    

 

Scholarly Papers in Nursing Research:

The literature is replete with examples of valid and reliable tools used in nursing research as well as examples of misapplication of psychometric testing. Reports of studies frequently included information about content validity. Criterion validity was rarely reported and errors in measurement of the criterion were identified. Construct validity remains under reported. Most researchers have reported internal consistency reliability (α), but few articles indicate any type of stability reliability testing. When retest reliability was asserted, time intervals and correlations were frequently not included (Holli A. DeVon, etal, 2007).   

 

CONCLUSION:

Lack of information on psychometric properties and misapplication of psychometric testing is common in the literature. Manuscripts are not usually rejected solely because of lack of information on reliability and validity. The strength of psychometric tests is dependent on many important factors such as underlying theory, the consequences of weak measures and the implications of the findings. Therefore, focus on rigid adherence to statistical rules and more emphasis on evaluating the strength of psychometric tests, planning for future studies and publishing data based articles are stressed.

 

REFERENCE:

1.       Carmines, E.G., and Zeller, R.A. (1979). Reliability and validity assessment. Thousand Oaks, CA: Sage

2.       Cronbach, L.J., and Meehl, P.E. (1955). Construct validity in psychological tests. Psychological Bulletin, 52, 281–302

3.       Ferketich, S. (1990). Focus on psychometrics: Internal consistency estimates of reliability. Research in Nursing and Health, 13, 437–440

4.       Grimm, G. L., and Yarnold, P. R. (2000). Reading and understanding more multivariate statistics. Washington, DC: American Psychological Association

5.       Holli A. DeVon, etal (2007). A Psychometric Toolbox for Testing Validity and Reliability. Journal of Nursing Scholarship Second Quarter  39:2, 155-164

6.       Lawshe, C.H. (1975). A quantitative approach to content validity. Personnel Psychology, 28, 563–575.

7.       Lynn, M.R. (1986). Determination and quantification of content validity. Nursing Research, 35, 382–385.

8.       Munro, B.H. (2005). Statistical methods for health care research (4th ed.).Philadelphia: Lippincott, Williams and Wilkins.

9.       Nunnally, J.C., and Bernstein, I.H. (1994). Psychometric theory (3rd ed.). New York: McGraw-Hill

10.     Polit, D.F., and Beck, C.T. (2006). The content validity index: Are you sure you know what’s being reported? Critique and recommendations. Research in Nursing and Health, 29, 489–497

11.     Waltz, C.F., Strickland, O.L., and Lenz, E.R. (2005). Measurement in nursing and health research (3rd ed.). New York: Springer.

 

 

 

Received on 19.06.2017           Modified on 18.07.2017

Accepted on 20.09.2017     © A&V Publications all right reserved