L. Dayle Yeager, Professor
Industrial Applied Research
Data are pieces of information that may be collected and assessed to reach decisions
regarding pertinent research questions for a given investigation.
As related to
hypotheses, such information may also be used to test hypotheses for either acceptance or
(It should be noted that hypotheses are never proved or disproved; they are
only accepted or rejected.)
Methods of data collection are often conducted by instruments that may be classified
according to several basic outcomes, namely: (1) surveying and (2) identifying.
instruments may be further categorized as either paper or mechanical instruments.
human behavior, traits, and perceptions are involved, paper instruments in the form of
questionnaires may be used to survey samples of populations.
Another paper instrument
involves data collected from various testing formats, including investigator-designed
exams and standardized exams.
Such exams may be used to identify achievement levels,
cognitive levels, perceptions, attitudes, and personality traits for students, employees,
The method of data compilation for identifying usually involves some form
of mechanical instrumentation to classify conditions associated with nonhuman activities.
Examples may include thermometers, dosimeters, digital measuring devices, etc. to
ascertain characteristics and qualities of samples selected from a nonhuman parameter.
should not be overlooked, however, that these mechanical means may also be used to
survey and identify certain human traits such as physical height, blood pressure,
Measuring Instrument Criteria
Even though there are various criteria used to judge the quality of measuring instruments,
validity and reliability remain as dominant characteristics.
Validity may be simply
defined as the ability of an instrument to measure what it purports to measure while
reliability addresses an instrument’s ability to consistency measure what it is intended to
As related to validity, various questions are as follow:
To what extent does this instrument represent the general
domain of interest?
b) Criterion-related validity:
To what extent does this instrument correlate with
Face validity: To what extent does this instrument appear to be able to measure
what it is intended to measure?
d) Predictive validity:
To what extent can the instrument’s measurement predict
future actions and/or conditions in similar, but delayed situations?
To what extent does this instrument measure what is
suppose to measure?